# ESA anemia correction × locoregional tumor control during head-and-neck radiotherapy — evidence checklist Verdict 2940 · Evidence Check Stored-material review date: Not recorded Original evidence cut-off: 2026-08-26 Review-criteria version: 1.0 Public page/material preparation date: 2026-10-11 Clinical papers were not read directly; existing review summaries were checked. No new clinical-literature search was performed. Unverified does not mean that research is absent or that there is no effect. ## Assessment scope - Population: Head-and-neck squamous-cell carcinoma radiotherapy; original DAHANCA inclusion of Hb below 14.0 g/dL; stage and detailed eligibility remain unverified - Ingredient, formulation and route: Original studies of ESA anemia correction during radiotherapy; individual trial conditions for darbepoetin, epoetin beta and epoetin alfa - Dose: Historical weekly darbepoetin exposure in DAHANCA; actual amount, route, target and adherence remain unverified. This is not medication, anticancer, anemia-correction or radiotherapy guidance. - Comparator: Original trial-specific controls: ENHANCE placebo, RTOG ESA added or not added, and radiotherapy alone in the review; specific concomitant treatments remain unverified - Measured outcome: Existing question of improved locoregional tumor control; original time to failure and five-year cumulative failure are lower-is-better and showed worsening. Survival and Hb are separate outcomes. - Duration: Original DAHANCA five-year cumulative outcomes and interim stopping; RTOG three- and five-year outcomes and follow-up and review-version context; detailed individual and endpoint follow-up remain unverified - Questions outside this scope: Converting Hb correction into improved tumor control or survival; Reversing failure-rate or HR direction into benefit; Adding RTOG follow-up or review participants as independent groups; Identical effects or harms in other cancers, populations or Hb targets, kidney disease or preterm-infant indications; Current FDA-approval certification or medication, radiotherapy, anemia-correction or anticancer-treatment instructions ## Checklist items ### N01 · Is the selected original question improved locoregional tumor control? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The selected existing question is improved locoregional tumor control. The parent describes harm from ESA anemia correction during head-and-neck radiotherapy. The question's improvement direction is distinguished from the observed worsening. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. The final count of independent claims remains undetermined. - Reason for missing information: The final count of independent claims remains undetermined. - Required materials and conditions: Link selected and parent claims, disease, exposure and failure direction - Materials to obtain first: Link selected and parent claims, disease, exposure and failure direction - Further research: Proposal: If the question actually changes, separately record its original wording and excluded scope. - Conditions for verification: Preserve both the original positive question and the actual direction of its original results. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: Is the selected original question improved locoregional tumor control? - Application exceptions: Do not arbitrarily rename the improvement question as a harm question or create a new verdict or split. - Remaining uncertainty: Do not arbitrarily rename the improvement question as a harm question or create a new verdict or split. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M03 §3.1, §3.2.1–3.2.4 question-clarification principles; detailed separation rules are our own design / M03 §3.2.4; our own extension boundary - Related sources: R01 ### N02 · Are the subclaim's grade F and note separate from parent F12 and axes? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: Selected grade F and the note that five-year failure worsened to 47% versus 34% are preserved. Subclaim score, status and axes are absent and not recorded. Parent F12, safety Warning and axes belong to the parent. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. The original materials contain no distinct subclaim score, status or axes. - Reason for missing information: The original materials contain no distinct subclaim score, status or axes. - Required materials and conditions: Locations of actual and absent subclaim fields and original parent values - Materials to obtain first: Locations of actual and absent subclaim fields and original parent values - Further research: Proposal: If subclaim fields are actually added, distinguish the original version from the new event. - Conditions for verification: Distinguish the actual subclaim grade F from the parent verdict's scope. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: Are the subclaim's grade F and note separate from parent F12 and axes? - Application exceptions: Do not uniformly assign D to subclaims or copy parent 12 points, RX, I1, E-, B2/C0 as distinct subclaim values. - Remaining uncertainty: Do not uniformly assign D to subclaims or copy parent 12 points, RX, I1, E-, B2/C0 as distinct subclaim values. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M03 §3.1, §3.2.1–3.2.4 question-clarification principles; detailed separation rules are our own design / M04 §4.5, §4.6.3; current material-conversion limits reflect review scope and our own design - Related sources: R01 ### N03 · Do repeated refutation, design limitations and cross-checking constitute current review? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The original parent describes F12 based on repeated refutation across three ESA trials, DAHANCA harm, early stopping at a planned interim analysis and open-label design. RX, cross-checking, true/false indications and empty values are historical records. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. No current RoB2/GRADE assessment, clinical reassessment or methodological certification was performed. - Reason for missing information: No current RoB2/GRADE assessment, clinical reassessment or methodological certification was performed. - Required materials and conditions: Korean and English wording, axes, verdict, corrections and methodological statements - Materials to obtain first: Korean and English wording, axes, verdict, corrections and methodological statements - Further research: Proposal: If new evidence is requested, separate the original version from the new update event. - Conditions for verification: Separate original history from incomplete current content review. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: Do repeated refutation, design limitations and cross-checking constitute current review? - Application exceptions: Do not inherit RX, the true large-hard-endpoint RCT indication or historical independence and cross-checking as current review or independence certification. - Remaining uncertainty: Do not inherit RX, the true large-hard-endpoint RCT indication or historical independence and cross-checking as current review or independence certification. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M04 §4.5, §4.6.3; current material-conversion limits reflect review scope and our own design / M08 §8.1–8.2.3; our own application - Related sources: R01, R02, R03, R04 ### N04 · Which materials were actually read, and which external materials remain unread? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The selection and complete preserved summary were read for production. External abstracts, full texts, registrations, plans, original FDA warnings and verdict 2925's page and links remain unread; access was not attempted. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. No search of current external contents, all recent literature or regulatory materials was performed. - Reason for missing information: No search of current external contents, all recent literature or regulatory materials was performed. - Required materials and conditions: Acquisition and reading scope for preserved summaries and each external material - Materials to obtain first: Acquisition and reading scope for preserved summaries and each external material - Further research: Proposal: If material and reading records are supplied, record only the actual level. - Conditions for verification: Keep summary reading and external unread status separate for each material. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: Which materials were actually read, and which external materials remain unread? - Application exceptions: Do not convert the presence of a URL or a statement about original-paper roles into current full-text reading, link verification or failed access. - Remaining uncertainty: Do not convert the presence of a URL or a statement about original-paper roles into current full-text reading, link verification or failed access. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M04 §4.5–4.6 material-collection and selection principles; separation of access and reading and hyperlink rules are our own design / M04 §4.5, §4.6.3; current material-conversion limits reflect review scope and our own design - Related sources: R01, R02, R03, R04 ### N05 · How are trials, the review and initial and follow-up reports linked? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: DAHANCA 10, ENHANCE, the same-trial initial and long-term RTOG 99-03 report group and the five-RCT review are distinguished. The reference to 2925 concerns another indication and is not a report of this cancer trial. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Unique trial and participant totals and review inclusion and overlap were not directly verified. - Reason for missing information: Unique trial and participant totals and review inclusion and overlap were not directly verified. - Required materials and conditions: Original bibliography, review inclusion, initial and long-term reports and participant relationships - Materials to obtain first: Original bibliography, review inclusion, initial and long-term reports and participant relationships - Further research: Proposal: If original lists are supplied, link each report to the same, different or uncertain study family. - Conditions for verification: Distinguish preserved report groups from undetermined unique totals and other indications. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: How are trials, the review and initial and follow-up reports linked? - Application exceptions: Do not add the review, original trials, two RTOG PMIDs or 2925 as new independent trials or participants. - Remaining uncertainty: Do not add the review, original trials, two RTOG PMIDs or 2925 as new independent trials or participants. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M04 §4.6.1–4.6.2 distinction between studies and reports; cohort and outcome linking is our own application / M04 §4.6.1–4.6.2, M15 §15.3.1; our own application - Related sources: R01, R02, R03, R04 ### N06 · What were DAHANCA's population and Hb eligibility conditions? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The setting is head-and-neck squamous-cell carcinoma radiotherapy. The original DAHANCA condition included patients with Hb below 14.0 g/dL. Detailed stage, exact age and site and baseline anemia cause are absent from the summary. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Detailed stage, site, age, anemia cause and baseline risk remain unverified. - Reason for missing information: Detailed stage, site, age, anemia cause and baseline risk remain unverified. - Required materials and conditions: Stage, site, inclusion and exclusion, Hb, anemia cause and baseline risk - Materials to obtain first: Stage, site, inclusion and exclusion, Hb, anemia cause and baseline risk - Further research: Proposal: If original population tables are supplied, distinguish the actual population from gaps for other cancers and goals. - Conditions for verification: Record original population and Hb eligibility along with gaps in detailed conditions. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What were DAHANCA's population and Hb eligibility conditions? - Application exceptions: Do not reuse the Hb eligibility threshold as an effect for all severe-anemia or other-cancer populations. - Remaining uncertainty: Do not reuse the Hb eligibility threshold as an effect for all severe-anemia or other-cancer populations. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M03 §3.2.1; detailed population conditions are our own design / M03 §3.2.1–3.2.4; dose, persistence and effect-modification details are our own design - Related sources: R01 ### N07 · What were darbepoetin exposure, goals and hematologic response in the trial? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The historical DAHANCA condition used weekly darbepoetin during radiotherapy. Exact dose, formulation, route, Hb target and stopping rules are absent from the summary. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Dose, route, actual adherence, achieved Hb and discontinuation details remain unverified. - Reason for missing information: Dose, route, actual adherence, achieved Hb and discontinuation details remain unverified. - Required materials and conditions: Nominal and actual drug exposure, Hb target, changes and discontinuation - Materials to obtain first: Nominal and actual drug exposure, Hb target, changes and discontinuation - Further research: Proposal: If original exposure records are supplied, record trial conditions and actual use and goals. - Conditions for verification: Distinguish historical drug conditions from gaps in direct exposure and targets. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What were darbepoetin exposure, goals and hematologic response in the trial? - Application exceptions: Do not write the historical weekly exposure as medication, anemia-correction, anticancer or radiotherapy-selection instructions. - Remaining uncertainty: Do not write the historical weekly exposure as medication, anemia-correction, anticancer or radiotherapy-selection instructions. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M03 §3.2.2–3.2.3; reverse-direction labeling and assessment adaptations are our own design / M03 §3.2.3, M08 §8.2.2–8.2.3; our own application - Related sources: R01 ### N08 · What were control radiotherapy and shared or concurrent cancer treatments? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The original comparison is radiotherapy with an ESA added versus control. The review describes radiotherapy plus EPO versus radiotherapy alone. Specific radiotherapy and other concomitant treatments in each trial are absent from the summary. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Concomitant treatment, radiotherapy conditions and their balance remain unverified for each trial. - Reason for missing information: Concomitant treatment, radiotherapy conditions and their balance remain unverified for each trial. - Required materials and conditions: Trial-specific controls, radiotherapy plans and shared or concomitant cancer and anemia treatment - Materials to obtain first: Trial-specific controls, radiotherapy plans and shared or concomitant cancer and anemia treatment - Further research: Proposal: If original comparison materials are supplied, distinguish the added drug from shared and other treatments. - Conditions for verification: Separate original comparison wording from gaps in detailed conditions. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What were control radiotherapy and shared or concurrent cancer treatments? - Application exceptions: Do not use the review's control label to certify that every trial had no concomitant treatment. - Remaining uncertainty: Do not use the review's control label to certify that every trial had no concomitant treatment. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M03 §3.2.2–3.2.3; reverse-direction labeling and assessment adaptations are our own design / M03 §3.2.3, M08 §8.2.2–8.2.3; our own application / M19 §19.2.2–19.2.4 design considerations; detailed confounding records are our own design - Related sources: R01, R02, R03, R04 ### N09 · Are planned 600, randomized 522 and evaluable 513 different stages? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: DAHANCA's planned 600, 522 randomized at stopping and 513 evaluable (254/259) are separately preserved. Planning, allocation and evaluation represent different denominators. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Reasons for differences between allocation and evaluation, original arm correspondence and endpoint-denominator details remain externally unread. - Reason for missing information: Reasons for differences between allocation and evaluation, original arm correspondence and endpoint-denominator details remain externally unread. - Required materials and conditions: Allocation and evaluation flow by arm, correspondence of 254/259 to arms, exclusions and missingness - Materials to obtain first: Allocation and evaluation flow by arm, correspondence of 254/259 to arms, exclusions and missingness - Further research: Proposal: If original flow diagrams are supplied, record stages and inclusion rules by arm. - Conditions for verification: Distinguish original stage-specific denominators from unverified flow. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: Are planned 600, randomized 522 and evaluable 513 different stages? - Application exceptions: Do not subtract counts to calculate attrition counts or rates or use planned 600 as the actual analysis population. - Remaining uncertainty: Do not subtract counts to calculate attrition counts or rates or use planned 600 as the actual analysis population. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptations are our own design / M08 §8.2.2, M15 §15.3.1; our own application - Related sources: R01 ### N10 · What are the locoregional-failure time, five-year values and HR? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The primary endpoint was time to locoregional failure. Original worsening results were five-year cumulative failure of 47% with darbepoetin versus 34% in controls, HR 1.53 (95% CI 1.16–2.02). - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Failure components and definition, population at risk, censoring and model remain unverified. - Reason for missing information: Failure components and definition, population at risk, censoring and model remain unverified. - Required materials and conditions: Failure definition, timepoint and time, denominator, HR model and interval - Materials to obtain first: Failure definition, timepoint and time, denominator, HR model and interval - Further research: Proposal: If original results and methods are supplied, link cumulative values to the time-to-event estimate. - Conditions for verification: Attribute the original endpoint, 47/34%, HR and 95% CI to the relevant comparison. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What are the locoregional-failure time, five-year values and HR? - Application exceptions: Do not calculate new risk differences, CIs or P values from original proportions or HRs. - Remaining uncertainty: Do not calculate new risk differences, CIs or P values from original proportions or HRs. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptations are our own design / M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptations are our own design / M15 §15.3.1–15.3.3; our own application - Related sources: R01 ### N11 · How should lower-is-better failure and the improvement question be read? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The original reader answer explicitly states that lower locoregional failure is better. Failure of 47% versus 34% is originally described as worsening, rather than improvement. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Detailed endpoint-specific success and failure adjudication rules remain unverified. - Reason for missing information: Detailed endpoint-specific success and failure adjudication rules remain unverified. - Required materials and conditions: Endpoint direction and relationship between question and result - Materials to obtain first: Endpoint direction and relationship between question and result - Further research: Proposal: If original definitions are added, record actual directions and interpretations of other endpoints. - Conditions for verification: Clearly distinguish the original improvement question from the observed worsening direction. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: How should lower-is-better failure and the improvement question be read? - Application exceptions: Do not turn increased failure into improved tumor control or a positive benefit. - Remaining uncertainty: Do not turn increased failure into improved tumor control or a positive benefit. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptations are our own design / M04 §4.6.3, M15 §15.3.1–15.3.3; status separation is our own design / M03 §3.2.4; our own extension boundary - Related sources: R01 ### N12 · Are the five-year cumulative proportion and time-to-failure HR the same measure? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: Five-year cumulative locoregional failure and the time-to-failure HR are separate expressions. The original HR is not converted into a simple five-year risk ratio or difference in cumulative percentages. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Cumulative estimation, proportional hazards, competing risks and adjustment remain unverified. - Reason for missing information: Cumulative estimation, proportional hazards, competing risks and adjustment remain unverified. - Required materials and conditions: Cumulative estimation, censoring, HR estimand, time and competing events - Materials to obtain first: Cumulative estimation, censoring, HR estimand, time and competing events - Further research: Proposal: If original statistical methods are supplied, distinguish definitions of cumulative estimation and hazard comparisons. - Conditions for verification: Distinguish different original measures and gaps in their models. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: Are the five-year cumulative proportion and time-to-failure HR the same measure? - Application exceptions: Do not convert 47/34 or the HR into a new risk ratio, effect size or individual probability. - Remaining uncertainty: Do not convert 47/34 or the HR into a new risk ratio, effect size or individual probability. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptations are our own design / M08 §8.2.2, M15 §15.3.1; our own application / M15 §15.3.1–15.3.3; our own application - Related sources: R01 ### N13 · What are the actual analysis and risk denominators for locoregional outcomes? Verification status: Unverified Reading scope: Summary read; external materials not read - Current findings: The original evaluable 513 (254/259) and five-year failure percentages are preserved. The five-year population at risk, follow-up, censoring and actual analysis denominators for each endpoint are absent from the summary. - Why unverified: Required direct details are absent from the preserved summary and remain unverified. External materials remain unread. Five-year risk population, event counts and endpoint-specific analyses and censoring remain unverified. - Reason for missing information: Five-year risk population, event counts and endpoint-specific analyses and censoring remain unverified. - Required materials and conditions: Endpoint-specific analysis and risk populations, events and people, censoring and missingness - Materials to obtain first: Endpoint-specific analysis and risk populations, events and people, censoring and missingness - Further research: Proposal: If original risk tables or flow diagrams are supplied, distinguish evaluation from timepoint and endpoint denominators. - Conditions for verification: Record the original evaluation size and gaps in direct outcome denominators. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What are the actual analysis and risk denominators for locoregional outcomes? - Application exceptions: Do not copy evaluable 513 into every timepoint, survival or safety denominator or back-calculate event counts. - Remaining uncertainty: Do not copy evaluable 513 into every timepoint, survival or safety denominator or back-calculate event counts. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptations are our own design / M08 §8.2.2, M15 §15.3.1; our own application - Related sources: R01 ### N14 · What was the actual scope of open-label design and central blinded review? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The original DAHANCA account describes open-label design and says locoregional failure was not uniformly adjudicated through central blinded imaging review for every patient. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Patient-specific adjudication pathways, partial central review and assessor and analyst roles remain unverified. - Reason for missing information: Patient-specific adjudication pathways, partial central review and assessor and analyst roles remain unverified. - Required materials and conditions: Failure diagnosis and adjudication, imaging and clinical pathways and blinding by role - Materials to obtain first: Failure diagnosis and adjudication, imaging and clinical pathways and blinding by role - Further research: Proposal: If original adjudication methods are supplied, record actual pathways and limitations by role. - Conditions for verification: Separate original open-label and review limitations from direct details. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What was the actual scope of open-label design and central blinded review? - Application exceptions: Do not expand the absence of uniform central blinded review into an absence of all adjudication or examinations. - Remaining uncertainty: Do not expand the absence of uniform central blinded review into an absence of all adjudication or examinations. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M08 Table 8.2.a, M03 §3.2.4; our own application / M08 §8.1–8.2.3; our own application - Related sources: R01 ### N15 · What is known about randomization, allocation concealment and baseline balance? Verification status: Unverified Reading scope: Summary read; external materials not read - Current findings: The original summary describes DAHANCA as a multicenter randomized phase 3 trial. Actual tables or methods for sequence generation, concealment, stratification, baseline and cancer-treatment balance are absent. - Why unverified: Required direct details are absent from the preserved summary and remain unverified. External materials remain unread. Direct methods for allocation, concealment and baseline balance remain unverified. - Reason for missing information: Direct methods for allocation, concealment and baseline balance remain unverified. - Required materials and conditions: Allocation procedures, concealment, baseline and stratification factors and treatment balance - Materials to obtain first: Allocation procedures, concealment, baseline and stratification factors and treatment balance - Further research: Proposal: If original methods and baseline data are supplied, distinguish design labels from actual procedures. - Conditions for verification: Separate original design from gaps in direct procedures. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What is known about randomization, allocation concealment and baseline balance? - Application exceptions: Do not certify fulfillment of all methodological conditions from the randomized label, B2 or a past verdict. - Remaining uncertainty: Do not certify fulfillment of all methodological conditions from the randomized label, B2 or a past verdict. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M08 §8.1–8.2.3; our own application / M03 §3.2.1; detailed population conditions are our own design - Related sources: R01 ### N16 · What does early stopping at the planned interim analysis preserve? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The original trial account states that inferiority of the experimental arm at a planned interim analysis led to stopping at 522, before the planned 600. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Interim thresholds, boundaries, multiplicity, committee and subsequent follow-up details remain unverified. - Reason for missing information: Interim thresholds, boundaries, multiplicity, committee and subsequent follow-up details remain unverified. - Required materials and conditions: Interim timing and thresholds, committee, alpha, stopping and subsequent follow-up - Materials to obtain first: Interim timing and thresholds, committee, alpha, stopping and subsequent follow-up - Further research: Proposal: If original plans and stopping records are supplied, record prior rules and actual stopping and follow-up. - Conditions for verification: Preserve the original stopping event and direct gaps in planning and follow-up. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What does early stopping at the planned interim analysis preserve? - Application exceptions: Do not turn the original planned-analysis wording into direct certification of advance planning or an exact bias magnitude. - Remaining uncertainty: Do not turn the original planned-analysis wording into direct certification of advance planning or an exact bias magnitude. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M08 Table 8.2.a, M04 §4.6.3; our own application / M08 §8.2.2, M15 §15.3.1; our own application / M15 §15.3.1–15.3.3; our own application - Related sources: R01 ### N17 · Does hematologic correction constitute a tumor-control benefit? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The original crux states that anemia measures improved but cancer control and survival worsened. Actual Hb changes, responders and target attainment are absent from the summary. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Quantitative Hb response, tumor oxygenation and mediation analyses remain unverified. - Reason for missing information: Quantitative Hb response, tumor oxygenation and mediation analyses remain unverified. - Required materials and conditions: Hematologic response, achieved Hb, denominator, tumor oxygenation and direct events - Materials to obtain first: Hematologic response, achieved Hb, denominator, tumor oxygenation and direct events - Further research: Proposal: If original response data are supplied, record Hb measures and clinical outcomes separately. - Conditions for verification: Preserve original hematologic statements separately from direct clinical worsening. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: Does hematologic correction constitute a tumor-control benefit? - Application exceptions: Do not treat increased Hb as a guarantee of oxygenation, tumor-control or survival improvement. - Remaining uncertainty: Do not treat increased Hb as a guarantee of oxygenation, tumor-control or survival improvement. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptations are our own design / M03 §3.2.4; our own extension boundary - Related sources: R01 ### N18 · What is the original event-free-survival value? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: DAHANCA event-free survival HR 1.36 (1.09–1.69) is described in the original harmful direction. This is a different outcome from the locoregional-failure HR. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. EFS components, model, denominator and detailed interval confidence level remain unverified. - Reason for missing information: EFS components, model, denominator and detailed interval confidence level remain unverified. - Required materials and conditions: EFS definition and components, denominator, HR model and interval confidence level - Materials to obtain first: EFS definition and components, denominator, HR model and interval confidence level - Further research: Proposal: If original results and methods are supplied, link EFS components and estimand to that value. - Conditions for verification: Keep the original EFS HR, interval and direction separate. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What is the original event-free-survival value? - Application exceptions: Do not automatically copy locoregional-failure values or its 95% CI designation into EFS. - Remaining uncertainty: Do not automatically copy locoregional-failure values or its 95% CI designation into EFS. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptations are our own design / M04 §4.6.3, M15 §15.3.1–15.3.3; status separation is our own design / M03 §3.2.4; our own extension boundary - Related sources: R01 ### N19 · What is the disease-specific-death result? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: DAHANCA disease-specific-death HR 1.43 (1.08–1.90) points in the original harmful direction. It is separate from overall death and failure. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Cause-of-death adjudication, events, risk population and detailed interval confidence level remain unverified. - Reason for missing information: Cause-of-death adjudication, events, risk population and detailed interval confidence level remain unverified. - Required materials and conditions: Cause-of-death adjudication, events and time, denominator, model and interval - Materials to obtain first: Cause-of-death adjudication, events and time, denominator, model and interval - Further research: Proposal: If original mortality outcomes are supplied, record causes, denominators and estimates separately. - Conditions for verification: Separate the original disease-specific HR from gaps in direct definitions. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What is the disease-specific-death result? - Application exceptions: Do not fill disease-specific death with another endpoint's HR or 95% designation. - Remaining uncertainty: Do not fill disease-specific death with another endpoint's HR or 95% designation. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptations are our own design / M04 §4.6.3, M15 §15.3.1–15.3.3; status separation is our own design / M03 §3.2.4; our own extension boundary - Related sources: R01 ### N20 · What is the original overall-death HR? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: DAHANCA overall-death HR 1.30 (1.02–1.64) is described in the original harmful direction. It is not mixed with numbers from other survival or failure outcomes. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Original overall-death count, risk population and detailed interval confidence level remain unverified. - Reason for missing information: Original overall-death count, risk population and detailed interval confidence level remain unverified. - Required materials and conditions: Overall death, risk population and denominator, time, model and interval - Materials to obtain first: Overall death, risk population and denominator, time, model and interval - Further research: Proposal: If original mortality tables are supplied, record the actual overall-death estimate, time and denominator. - Conditions for verification: Preserve original overall-death values and gaps. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What is the original overall-death HR? - Application exceptions: Do not calculate death counts, absolute-risk differences or new intervals from the original HR. - Remaining uncertainty: Do not calculate death counts, absolute-risk differences or new intervals from the original HR. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptations are our own design / M04 §4.6.3, M15 §15.3.1–15.3.3; status separation is our own design / M03 §3.2.4; our own extension boundary - Related sources: R01 ### N21 · Is the sibling survival-improvement question separate from the main tumor question? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The sibling survival question is improved survival, grade F, with a note about harmful disease-specific and overall death. It is preserved as context and remains separate from the selected locoregional question. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. This record also contains no distinct score, status or axes for the sibling. - Reason for missing information: This record also contains no distinct score, status or axes for the sibling. - Required materials and conditions: Existing sibling question, endpoint definitions and denominators and original-result links - Materials to obtain first: Existing sibling question, endpoint definitions and denominators and original-result links - Further research: Proposal: If the original question changes, separately record each existing question and its links. - Conditions for verification: Distinguish original survival context from the selected question. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: Is the sibling survival-improvement question separate from the main tumor question? - Application exceptions: Do not count the sibling row as a new trial or participant group or copy failure values into survival outcomes. - Remaining uncertainty: Do not count the sibling row as a new trial or participant group or copy failure values into survival outcomes. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M03 §3.1, §3.2.1–3.2.4 question-clarification principles; detailed separation rules are our own design / M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptations are our own design / M03 §3.2.4; our own extension boundary - Related sources: R01 ### N22 · What are the details of follow-up, treatment deviations, missingness and censoring? Verification status: Unverified Reading scope: Summary read; external materials not read - Current findings: Evaluable 513, five-year cumulative results and early stopping are preserved. Individual follow-up, treatment deviation or crossover, exclusions, censoring and ITT/PP details are absent from the summary. - Why unverified: Required direct details are absent from the preserved summary and remain unverified. External materials remain unread. Individual follow-up and deviations, ITT/PP and censoring and missing-data handling details remain unverified. - Reason for missing information: Individual follow-up and deviations, ITT/PP and censoring and missing-data handling details remain unverified. - Required materials and conditions: Endpoint-specific flow, follow-up and censoring, treatment deviations and missing-data handling - Materials to obtain first: Endpoint-specific flow, follow-up and censoring, treatment deviations and missing-data handling - Further research: Proposal: If original flow and analyses are supplied, distinguish allocation, evaluation, risk populations and losses. - Conditions for verification: Record direct gaps in original flow and time. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What are the details of follow-up, treatment deviations, missingness and censoring? - Application exceptions: Do not calculate or infer new attrition or analysis rules from differences in allocation and evaluation counts. - Remaining uncertainty: Do not calculate or infer new attrition or analysis rules from differences in allocation and evaluation counts. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptations are our own design / M08 §8.2.2, M15 §15.3.1; our own application / M19 §19.1.2–19.2.4, M04 §4.6.3; our own application - Related sources: R01 ### N23 · What history underlies the original registration-absence statements? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The original DAHANCA and ENHANCE registration values say absent because they predate registration requirements; RTOG records NCT00004917, and the review says not applicable. These original expressions are preserved; external registration existence was not currently verified. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. External registration existence and prior plans remain unread. - Reason for missing information: External registration existence and prior plans remain unread. - Required materials and conditions: Original registration, timing, contemporaneous records and plans and changes - Materials to obtain first: Original registration, timing, contemporaneous records and plans and changes - Further research: Proposal: If original registration and planning materials are supplied, distinguish actual existence and contents from original history. - Conditions for verification: Distinguish original values from the current unqueried scope. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What history underlies the original registration-absence statements? - Application exceptions: Do not turn the historical absence wording into current certification of exhaustive registry absence or a new identifier. - Remaining uncertainty: Do not turn the historical absence wording into current certification of exhaustive registry absence or a new identifier. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M04 §4.5–4.6 material-collection and selection principles; separation of access and reading and hyperlink rules are our own design / M04 §4.5, §4.6.3; current material-conversion limits reflect review scope and our own design / M08 Table 8.2.a, M04 §4.6.3; our own application - Related sources: R01, R02, R03, R04 ### N24 · Were prior endpoints, interim plans and the hierarchy of multiple outcomes read? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The original account includes primary failure, a planned interim analysis and major survival outcomes. Protocols, SAPs, all secondary and exploratory hierarchies, multiplicity and changes remain unread. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Direct documents for prior hierarchies, testing and changes remain unverified. - Reason for missing information: Direct documents for prior hierarchies, testing and changes remain unverified. - Required materials and conditions: Original plans, prior hierarchy, interim and final testing and changes - Materials to obtain first: Original plans, prior hierarchy, interim and final testing and changes - Further research: Proposal: If original plans are supplied, distinguish contemporaneous roles from reports and follow-up. - Conditions for verification: Separate original roles from gaps in planning details. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: Were prior endpoints, interim plans and the hierarchy of multiple outcomes read? - Application exceptions: Do not certify every prior definition solely from the original primary-endpoint or planned-analysis labels. - Remaining uncertainty: Do not certify every prior definition solely from the original primary-endpoint or planned-analysis labels. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M08 Table 8.2.a, M04 §4.6.3; our own application / M04 §4.5–4.6 material-collection and selection principles; separation of access and reading and hyperlink rules are our own design - Related sources: R01 ### N25 · What are the clinical thresholds, individual effects and subgroups? Verification status: Unverified Reading scope: Summary read; external materials not read - Current findings: Original failure and HR intervals and harmful-survival statements are available. Clinical-success thresholds, individual-response criteria and Hb, stage or population interactions are absent from the summary. - Why unverified: Required direct details are absent from the preserved summary and remain unverified. External materials remain unread. Prior clinical thresholds and direct subgroup effects and denominators remain unverified. - Reason for missing information: Prior clinical thresholds and direct subgroup effects and denominators remain unverified. - Required materials and conditions: Prior importance thresholds and subgroup hypotheses, interactions, denominators and intervals - Materials to obtain first: Prior importance thresholds and subgroup hypotheses, interactions, denominators and intervals - Further research: Proposal: If original plans or subgroup data are supplied, distinguish overall estimates from separate conditions. - Conditions for verification: Record overall original results and gaps in subgroup effects and clinical importance. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What are the clinical thresholds, individual effects and subgroups? - Application exceptions: Do not copy original effects into every stage, Hb target, dose or individual outcome. - Remaining uncertainty: Do not copy original effects into every stage, Hb target, dose or individual outcome. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M15 §15.3.1–15.3.3; our own application / M03 §3.2.1–3.2.4; dose, persistence and effect-modification details are our own design - Related sources: R01 ### N26 · What are ENHANCE's original population, design and RR results? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The original ENHANCE notation is 351 patients (epoetin beta 180, placebo 171), ITT. Primary locoregional progression-free survival adjusted RR 1.62 (95% CI 1.22–2.14), P=.0008, and survival RR 1.39 (1.05–1.84) point in the original harmful direction. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Adjustment variables, actual endpoints and ITT handling and detailed survival-interval confidence level remain unverified. - Reason for missing information: Adjustment variables, actual endpoints and ITT handling and detailed survival-interval confidence level remain unverified. - Required materials and conditions: PFS and survival definitions, adjusted model, actual denominators, ITT and intervals - Materials to obtain first: PFS and survival definitions, adjusted model, actual denominators, ITT and intervals - Further research: Proposal: If original ENHANCE methods and tables are supplied, record RR notation, estimand and endpoint-specific analyses. - Conditions for verification: Preserve each original RR, interval and P value and trial-specific denominator and hierarchy. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What are ENHANCE's original population, design and RR results? - Application exceptions: Do not change original RRs into DAHANCA HRs or probabilities of improvement. - Remaining uncertainty: Do not change original RRs into DAHANCA HRs or probabilities of improvement. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptations are our own design / M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptations are our own design / M04 §4.6.3, M15 §15.3.1–15.3.3; status separation is our own design / M08 §8.1–8.2.3; our own application - Related sources: R02 ### N27 · How do epoetin beta, epoetin alfa and darbepoetin exposures differ? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The original verdict concerns three ESAs in the same head-and-neck radiotherapy setting, rather than one drug alone. Trial-specific doses, Hb targets, actual exposures and combinations are not detailed in the summary. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Direct details of trial-specific targets, exposures, combinations and actual Hb remain unverified. - Reason for missing information: Direct details of trial-specific targets, exposures, combinations and actual Hb remain unverified. - Required materials and conditions: Drug-specific formulation, amount and targets, shared and different treatments and populations - Materials to obtain first: Drug-specific formulation, amount and targets, shared and different treatments and populations - Further research: Proposal: If original intervention data are supplied, record drugs, targets and trial-specific conditions separately. - Conditions for verification: Record shared claim context together with actual drug and exposure differences. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: How do epoetin beta, epoetin alfa and darbepoetin exposures differ? - Application exceptions: Do not fill different drugs or targets with the same dose or conditions or extend them to every other indication. - Remaining uncertainty: Do not fill different drugs or targets with the same dose or conditions or extend them to every other indication. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M03 §3.2.2–3.2.3; reverse-direction labeling and assessment adaptations are our own design / M03 §3.2.3, M08 §8.2.2–8.2.3; our own application / M03 §3.2.1–3.2.4; dose, persistence and effect-modification details are our own design - Related sources: R01, R02, R03 ### N28 · What are RTOG's initial and long-term endpoint values? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: RTOG's 148 patients are described as showing no epoetin alfa benefit. Three-year failure 44% versus 36% (P=.56), five-year failure 46.2% versus 39.4% (P=.42) and OS 36.9% versus 38.2% (P=.54) are separately preserved. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Actual analysis denominators, original CIs and models and detailed OS timepoint definitions remain unverified. - Reason for missing information: Actual analysis denominators, original CIs and models and detailed OS timepoint definitions remain unverified. - Required materials and conditions: Timepoint-specific failure and OS, denominators, estimates, CIs and analysis plans - Materials to obtain first: Timepoint-specific failure and OS, denominators, estimates, CIs and analysis plans - Further research: Proposal: If original RTOG tables are supplied, link each timepoint and endpoint to its original intervals and denominators. - Conditions for verification: Keep original timepoints, proportions and P values attached to their respective endpoints. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What are RTOG's initial and long-term endpoint values? - Application exceptions: Do not turn nonsignificance into zero effect, equivalence or guaranteed individual survival. - Remaining uncertainty: Do not turn nonsignificance into zero effect, equivalence or guaranteed individual survival. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptations are our own design / M04 §4.6.3, M15 §15.3.1–15.3.3; status separation is our own design / M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptations are our own design - Related sources: R03 ### N29 · Do the two RTOG PMIDs refer to reports of the same trial? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The RTOG citation preserves initial PMID 17716826 and long-term PMID 25670542 together. Its original flag groups reports of the same trial as one. The URL identifies the long-term report's bibliography. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Actual follow-up and denominators of the follow-up report and each original full text remain unread. - Reason for missing information: Actual follow-up and denominators of the follow-up report and each original full text remain unread. - Required materials and conditions: Initial and follow-up populations, timepoints and analysis and report relationships - Materials to obtain first: Initial and follow-up populations, timepoints and analysis and report relationships - Further research: Proposal: If original reports are supplied, distinguish same-cohort follow-up from separate trials. - Conditions for verification: Preserve the original report group, bibliography and same-population relationship. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: Do the two RTOG PMIDs refer to reports of the same trial? - Application exceptions: Do not add two PMIDs or timepoints as two groups of 148 participants or independent replications. - Remaining uncertainty: Do not add two PMIDs or timepoints as two groups of 148 participants or independent replications. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M04 §4.6.1–4.6.2 distinction between studies and reports; cohort and outcome linking is our own application / M04 §4.6.1–4.6.2, M15 §15.3.1; our own application - Related sources: R03 ### N30 · Can the five-RCT, 1,397-participant review be added to original trials? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The original review size is five RCTs and 1,397 participants. Report and included-trial lists and overlap with current original trials and follow-up were not directly read. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Review inclusion, contributions and unique trial and participant totals remain undetermined. - Reason for missing information: Review inclusion, contributions and unique trial and participant totals remain undetermined. - Required materials and conditions: Inclusion and contribution lists, trial identities, populations and timepoints and overlap - Materials to obtain first: Inclusion and contribution lists, trial identities, populations and timepoints and overlap - Further research: Proposal: If original inclusion tables are supplied, link the review to unique original study families. - Conditions for verification: Distinguish original review counts from undetermined unique totals. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: Can the five-RCT, 1,397-participant review be added to original trials? - Application exceptions: Do not add review 1,397 and individual trials or follow-up to create new participant or trial totals. - Remaining uncertainty: Do not add review 1,397 and individual trials or follow-up to create new participant or trial totals. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M04 §4.6.1–4.6.2 distinction between studies and reports; cohort and outcome linking is our own application / M04 §4.6.1–4.6.2, M15 §15.3.1; our own application / M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptations are our own design - Related sources: R04 ### N31 · What is the event direction of the review's ORs? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: OS Peto OR 0.73 (95% CI 0.58–0.91) and locoregional PFS OR 0.63 (0.49–0.80) are originally described as worse with radiotherapy plus EPO and favoring radiotherapy alone. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. OR coding, synthesis model and detailed PFS-interval confidence level remain unverified. - Reason for missing information: OR coding, synthesis model and detailed PFS-interval confidence level remain unverified. - Required materials and conditions: Review event and success coding, models and endpoint-specific contributions and intervals - Materials to obtain first: Review event and success coding, models and endpoint-specific contributions and intervals - Further research: Proposal: If original review tables are supplied, record OR coding, estimands and contributions. - Conditions for verification: Preserve the original review direction, measures and intervals at their stated level. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What is the event direction of the review's ORs? - Application exceptions: Do not reverse OR<1 into reduced death or failure with ESA or convert it into an HR or RR. - Remaining uncertainty: Do not reverse OR<1 into reduced death or failure with ESA or convert it into an HR or RR. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptations are our own design / M04 §4.6.3, M15 §15.3.1–15.3.3; status separation is our own design / M15 §15.3.1–15.3.3; our own application - Related sources: R04 ### N32 · What are the review's timing, subsets, heterogeneity and direct comparisons? Verification status: Unverified Reading scope: Summary read; external materials not read - Current findings: The review's original bibliography is from 2009, whereas the original DAHANCA report is from 2018. Actual inclusion and updates, contributions by drug, target and timepoint, heterogeneity and subgroup analyses are absent from the summary. - Why unverified: Required direct details are absent from the preserved summary and remain unverified. External materials remain unread. Temporal relationships between versions and inclusion, subsets and quantitative heterogeneity remain unverified. - Reason for missing information: Temporal relationships between versions and inclusion, subsets and quantitative heterogeneity remain unverified. - Required materials and conditions: Review version, inclusion timing, endpoint subsets, synthesis, heterogeneity and conditions - Materials to obtain first: Review version, inclusion timing, endpoint subsets, synthesis, heterogeneity and conditions - Further research: Proposal: If original review tables are supplied, record versions and actual included families and conditions. - Conditions for verification: Preserve original dates and bibliography and gaps in actual inclusion. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What are the review's timing, subsets, heterogeneity and direct comparisons? - Application exceptions: Do not assume every recent report was included in the review or arbitrarily correct years or identities. - Remaining uncertainty: Do not assume every recent report was included in the review or arbitrarily correct years or identities. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M04 §4.6.1–4.6.2, M15 §15.3.1; our own application / M08 §8.2.2, M15 §15.3.1; our own application / M03 §3.2.1–3.2.4; dose, persistence and effect-modification details are our own design - Related sources: R04 ### N33 · Are repeated refutation in the same indication and other goals distinguished? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The original scope is limited to improved tumor control during cancer radiotherapy. Kidney-disease anemia and transfusion reduction in very-low-birth-weight preterm infants, verdict 2925 B72, concern different populations and goals. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Pages and original materials for other indications remain unread. - Reason for missing information: Pages and original materials for other indications remain unread. - Required materials and conditions: Direct questions, exposures and endpoints for each indication and goal - Materials to obtain first: Direct questions, exposures and endpoints for each indication and goal - Further research: Proposal: If materials for another requested question are supplied, record its original unit and population separately. - Conditions for verification: Preserve only the original link context for another goal and do not create treatment-selection instructions. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: Are repeated refutation in the same indication and other goals distinguished? - Application exceptions: Do not merge 2925 as tumor-control benefit, safety evidence or a separate anticancer trial here. - Remaining uncertainty: Do not merge 2925 as tumor-control benefit, safety evidence or a separate anticancer trial here. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M03 §3.2.1; detailed population conditions are our own design / M03 §3.2.2–3.2.3; reverse-direction labeling and assessment adaptations are our own design / M03 §3.2.4; our own extension boundary - Related sources: R01, R02, R03, R04 ### N34 · What is the original statement of no differences in ancillary safety? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The original counterview says no excess serious adverse events, cardiovascular increase or differences in acute or late radiation toxicity were identified. Quantitative events, denominators, CIs and P values are absent from the summary. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Direct safety event counts, intervals and collection methods remain unverified. - Reason for missing information: Direct safety event counts, intervals and collection methods remain unverified. - Required materials and conditions: Safety definitions and collection, individual events and denominators, estimates and intervals - Materials to obtain first: Safety definitions and collection, individual events and denominators, estimates and intervals - Further research: Proposal: If original safety tables are supplied, record original observations and intervals for each item. - Conditions for verification: Preserve both original ancillary-safety statements and direct clinical worsening. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What is the original statement of no differences in ancillary safety? - Application exceptions: Do not turn unverified differences into zero events, equivalence of all safety outcomes or harmlessness. - Remaining uncertainty: Do not turn unverified differences into zero events, equivalence of all safety outcomes or harmlessness. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M19 §19.1.2–19.2.4; detailed distinctions are our own design / M19 §19.2.2–19.2.4, M03 §3.2.3; our own application / M19 §19.1.2–19.2.4; our own application / M04 §4.6.3, M15 §15.3.1–15.3.3; status separation is our own design - Related sources: R01 ### N35 · Are major tumor and survival harms separate from safety collection? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The original explanation states that demonstrated harm came from primary and major tumor-control and survival endpoints, rather than ancillary safety items. A complete list of all adverse events is absent from the summary. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Complete harm definitions, collection, discontinuation and severity tables remain unverified. - Reason for missing information: Complete harm definitions, collection, discontinuation and severity tables remain unverified. - Required materials and conditions: Clinical-harm and safety endpoints, collection and adjudication, severity and discontinuation - Materials to obtain first: Clinical-harm and safety endpoints, collection and adjudication, severity and discontinuation - Further research: Proposal: If original safety and endpoint data are supplied, distinguish actual clinical outcomes from separate collection. - Conditions for verification: Distinguish endpoint-specific harm from gaps in complete collection. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: Are major tumor and survival harms separate from safety collection? - Application exceptions: Do not erase demonstrated tumor or survival harm based on no differences in ancillary safety. - Remaining uncertainty: Do not erase demonstrated tumor or survival harm based on no differences in ancillary safety. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M19 §19.1.2–19.2.4; detailed distinctions are our own design / CHAMGAP's own interpretation boundary. M19 principles are referenced; no official causality tool was run / M19 §19.1.2–19.2.4; our own application - Related sources: R01 ### N36 · What are harm-event, participant and time denominators and timepoints? Verification status: Unverified Reading scope: Summary read; external materials not read - Current findings: Original context includes evaluable 513, five-year tumor outcomes and interim stopping. Numbers and time at risk for each safety event, onset, recovery, exposure and discontinuation timing are absent. - Why unverified: Required direct details are absent from the preserved summary and remain unverified. External materials remain unread. Denominators, time, discontinuation and recovery for each safety event remain unverified. - Reason for missing information: Denominators, time, discontinuation and recovery for each safety event remain unverified. - Required materials and conditions: Safety-specific denominators and durations, event timing and treatment and follow-up discontinuation - Materials to obtain first: Safety-specific denominators and durations, event timing and treatment and follow-up discontinuation - Further research: Proposal: If original safety flow is supplied, record efficacy evaluation separately from safety exposure and time. - Conditions for verification: Separate original clinical context from gaps in safety time and denominators. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What are harm-event, participant and time denominators and timepoints? - Application exceptions: Do not copy 513/522 or five-year figures as actual denominators or duration for every harm. - Remaining uncertainty: Do not copy 513/522 or five-year figures as actual denominators or duration for every harm. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M19 §19.2 design-specific harm questions; temporal-relationship details are our own design / M19 §19.1.2–19.2.4; our own application / M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptations are our own design - Related sources: R01 ### N37 · What is known about baseline cancer risk, concurrent treatments and alternative attribution? Verification status: Unverified Reading scope: Summary read; external materials not read - Current findings: Radiotherapy and ESA comparisons and clinical worsening are original-summary observations. Actual event-specific details of baseline and concurrent treatments, disease characteristics, other exposures and adjustment are absent. - Why unverified: Required direct details are absent from the preserved summary and remain unverified. External materials remain unread. Individual-event alternatives, mechanisms, concurrent treatments and adjustment details remain unverified. - Reason for missing information: Individual-event alternatives, mechanisms, concurrent treatments and adjustment details remain unverified. - Required materials and conditions: Baseline and concurrent exposures, event attribution, adjustment and alternative explanations - Materials to obtain first: Baseline and concurrent exposures, event attribution, adjustment and alternative explanations - Further research: Proposal: If original methods and event data are supplied, distinguish randomized comparisons from individual causes and mechanisms. - Conditions for verification: Separate original comparisons from gaps in attribution and mechanisms. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What is known about baseline cancer risk, concurrent treatments and alternative attribution? - Application exceptions: Do not turn population-level harmful direction into established causes or mechanisms for individual patients. - Remaining uncertainty: Do not turn population-level harmful direction into established causes or mechanisms for individual patients. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M19 §19.2.2–19.2.4 design considerations; detailed confounding records are our own design / M03 §3.2.3, M08 §8.2.2–8.2.3; our own application - Related sources: R01 ### N38 · What is known about missing and complete safety and clinical follow-up? Verification status: Unverified Reading scope: Summary read; external materials not read - Current findings: Original history includes evaluation and allocation, early stopping and open-label design. Safety surveillance, follow-up losses, omissions, censoring and completeness of all event reporting are absent from the summary. - Why unverified: Required direct details are absent from the preserved summary and remain unverified. External materials remain unread. Safety follow-up, omissions and overall reporting completeness remain unverified. - Reason for missing information: Safety follow-up, omissions and overall reporting completeness remain unverified. - Required materials and conditions: Safety and clinical flow, losses and omissions, reporting and censoring and sensitivity - Materials to obtain first: Safety and clinical flow, losses and omissions, reporting and censoring and sensitivity - Further research: Proposal: If original flow and surveillance records are supplied, distinguish observed outcomes from uncaptured and missing data. - Conditions for verification: Record original outcome limitations in follow-up and reporting. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What is known about missing and complete safety and clinical follow-up? - Application exceptions: Do not turn allocation and evaluation differences into absence of safety events or an exact bias magnitude. - Remaining uncertainty: Do not turn allocation and evaluation differences into absence of safety events or an exact bias magnitude. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M19 §19.1.2–19.2.4, M04 §4.6.3; our own application / M08 §8.2.2, M15 §15.3.1; our own application - Related sources: R01 ### N39 · What is the scope of FDA warnings, original harm, current verification and causality? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: Original advertising interpretation and safety explanation preserve the FDA boxed warning about shortened cancer survival and tumor progression or recurrence. The preserved summary contains no regulatory-original URL or separate citation. Current approvals and warning contents were not queried. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Official warning and approval texts, latest versions and direct mechanistic verification remain unread. - Reason for missing information: Official warning and approval texts, latest versions and direct mechanistic verification remain unread. - Required materials and conditions: Original regulatory text and versions, indications and clinical-outcome and judgment scope - Materials to obtain first: Original regulatory text and versions, indications and clinical-outcome and judgment scope - Further research: Proposal: If original regulatory materials are supplied, distinguish historical statements from direct contents and current updates. - Conditions for verification: Separate original warnings and clinical harm from the current unqueried regulatory scope. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What is the scope of FDA warnings, original harm, current verification and causality? - Application exceptions: Do not create new regulatory links, certify current approval or provide treatment, medication, radiotherapy-selection or dosing instructions. - Remaining uncertainty: Do not create new regulatory links, certify current approval or provide treatment, medication, radiotherapy-selection or dosing instructions. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: CHAMGAP's own interpretation boundary. M19 principles are referenced; no official causality tool was run / M04 §4.5–4.6 material-collection and selection principles; separation of access and reading and hyperlink rules are our own design / M04 §4.5, §4.6.3; current material-conversion limits reflect review scope and our own design - Related sources: R01 ### N40 · Does this answer questions about other cancers, Hb targets, populations or long-term and rare harms? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The original verdict concerns tumor control during head-and-neck radiotherapy. Direct results for other cancers, populations, Hb targets, other ESA exposures or rare and other long-term harms are absent from this summary. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Direct evidence for other populations, goals and rare harms remains unverified. - Reason for missing information: Direct evidence for other populations, goals and rare harms remains unverified. - Required materials and conditions: Direct population-, target- and exposure-specific materials and safety follow-up - Materials to obtain first: Direct population-, target- and exposure-specific materials and safety follow-up - Further research: Proposal: If direct materials are supplied, distinguish the selected claim from other indication and harm questions. - Conditions for verification: Record original limits and gaps in safety and effect extensions. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: Does this answer questions about other cancers, Hb targets, populations or long-term and rare harms? - Application exceptions: Do not extend original repeated refutation to identical harm for every ESA indication and exposure. - Remaining uncertainty: Do not extend original repeated refutation to identical harm for every ESA indication and exposure. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M19 §19.1.2–19.2.4; our own application / M03 §3.2.1–3.2.4; dose, persistence and effect-modification details are our own design / M03 §3.2.1; detailed population conditions are our own design - Related sources: R01 ### N41 · What were DAHANCA's mixed support, device or drug supply and roles? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: Danish Cancer Society and Amgen support, Amgen's free trial drug/Aranesp and main GCP and quality-control costs are preserved. The original account says the trial was investigator driven, Amgen had no design, conduct, collection or analysis role and offered no comments despite a manuscript-comment opportunity. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Original contracts, complete declarations and direct current role comparisons remain unread. - Reason for missing information: Original contracts, complete declarations and direct current role comparisons remain unread. - Required materials and conditions: Original sponsorship and supply, roles, contracts and data access and COI declarations - Materials to obtain first: Original sponsorship and supply, roles, contracts and data access and COI declarations - Further research: Proposal: If original declarations are supplied, distinguish support and supply from each actual operational role. - Conditions for verification: Separate original support, supply and role context from currently unread declarations. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What were DAHANCA's mixed support, device or drug supply and roles? - Application exceptions: Do not remove free-drug/GCP-cost or noninvolvement statements or upgrade I1 to current independence certification. - Remaining uncertainty: Do not remove free-drug/GCP-cost or noninvolvement statements or upgrade I1 to current independence certification. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: CHAMGAP's own independence and attribution conditions. Official methodological sections used in this pilot do not guarantee every detail of this item / M04 §4.6.1–4.6.2, M15 §15.3.1; our own application - Related sources: R01 ### N42 · What are funding and author relationships in other trials and the review? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: Original accounts record F. Hoffmann-La Roche support for ENHANCE, NCI/Ortho Biotech support for RTOG and mixed public and company funding of the review's included trials. Detailed company and investigator roles and the review's own declarations are absent. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Detailed roles, the review's own funding and COIs and investigator and population overlap remain unverified. - Reason for missing information: Detailed roles, the review's own funding and COIs and investigator and population overlap remain unverified. - Required materials and conditions: Report-specific sponsorship and supply, author, analysis and manuscript roles and review declarations - Materials to obtain first: Report-specific sponsorship and supply, author, analysis and manuscript roles and review declarations - Further research: Proposal: If original declarations and inclusion tables are supplied, record roles, families and funding by report. - Conditions for verification: Preserve report-specific original funding and relationship gaps. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: What are funding and author relationships in other trials and the review? - Application exceptions: Do not certify current investigator or population independence solely from different supporting institutions. - Remaining uncertainty: Do not certify current investigator or population independence solely from different supporting institutions. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: CHAMGAP's own independence and attribution conditions. Official methodological sections used in this pilot do not guarantee every detail of this item / M04 §4.6.1–4.6.2 distinction between studies and reports; cohort and outcome linking is our own application - Related sources: R02, R03, R04 ### N43 · Are original repeated refutation and unique trial and repeated-harm counts established? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: The original parent describes repeated refutation/RX across three different drugs and independent trials. This is preserved, but review inclusion, trial identities, replications and follow-up were not exhaustively compared, so unique counts and independence are not certified. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Unique trial totals, participant and event sums and independent repeated-harm counts remain undetermined. - Reason for missing information: Unique trial totals, participant and event sums and independent repeated-harm counts remain undetermined. - Required materials and conditions: Unique families, inclusion and follow-up, investigators and funding and event and population lists - Materials to obtain first: Unique families, inclusion and follow-up, investigators and funding and event and population lists - Further research: Proposal: If original lists are supplied, link same, different and uncertain families and repeated citations. - Conditions for verification: Distinguish original history from unrecorded, undetermined unique counts. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: Are original repeated refutation and unique trial and repeated-harm counts established? - Application exceptions: Do not add review, follow-up, the same population or endpoints as new independent trials or repeated events. - Remaining uncertainty: Do not add review, follow-up, the same population or endpoints as new independent trials or repeated events. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M04 §4.6.1–4.6.2 distinction between studies and reports; cohort and outcome linking is our own application / M04 §4.6.1–4.6.2, M15 §15.3.1; our own application / M04 §4.6.1–4.6.2; harm-material adaptations are our own design - Related sources: R01, R02, R03, R04 ### N44 · Are original dates, corrections, Korean and English texts and empty values preserved? Verification status: Partly verified Reading scope: Summary read; external materials not read - Current findings: Date 2026-08-26, empty corrections and deliberation, Korean and English texts and parent-axis true, false, unrecorded and empty-string values are preserved. - Why unverified: Original preserved-summary statements were linked for production as partly verified; clinical verification and content review are not complete. Separate evidence-cutoff and facet fields are absent; no latest correction, retraction or regulatory search was performed. - Reason for missing information: Separate evidence-cutoff and facet fields are absent; no latest correction, retraction or regulatory search was performed. - Required materials and conditions: Original dates, corrections and languages and current input and update scope - Materials to obtain first: Original dates, corrections and languages and current input and update scope - Further research: Proposal: If actual update materials are supplied, distinguish the complete original version from the correction event. - Conditions for verification: Record original history together with current production and incomplete content review. - Why this condition is needed: This condition is needed to record original information and gaps separately for the question: Are original dates, corrections, Korean and English texts and empty values preserved? - Application exceptions: Do not borrow historical dates or cross-checking as current clinical-verification dates or certify exhaustive absence from empty corrections. - Remaining uncertainty: Do not borrow historical dates or cross-checking as current clinical-verification dates or certify exhaustive absence from empty corrections. Existing content review remains incomplete and external original materials were not directly verified. - Referenced methods: M04 §4.5, §4.6.3; current material-conversion limits reflect review scope and our own design / M03 §3.1, §3.2.1–3.2.4 question-clarification principles; detailed separation rules are our own design - Related sources: R01, R02, R03, R04 ## Evidence and sources ### R01 Overgaard J, Hoff CM, Hansen HS, et al. DAHANCA 10—Effect of darbepoetin alfa and radiotherapy in squamous cell carcinoma of the head and neck. Radiother Oncol. 2018;127:12-19. PMID: 29523409. [PubMed](https://pubmed.ncbi.nlm.nih.gov/29523409/) Materials actually read: Summary read; external abstracts and full texts not read Findings recorded in the summary: 522 randomized/513 evaluable (254/259). Five-year cumulative failure 47% versus 34%, HR 1.53 (95% CI 1.16–2.02); EFS HR 1.36 (1.09–1.69), disease-specific death 1.43 (1.08–1.90) and overall death 1.30 (1.02–1.64) point in the original harmful direction. Original history records interim stopping at 522 of planned 600. Record level: Not recorded Independent replication: DAHANCA's same population, failure, survival, Hb, safety, interim and follow-up context; original independence statements are historical and do not currently certify unique families. ENHANCE's original drug, population, locoregional PFS and survival context is not added to other trial values or denominators. The two initial and long-term RTOG 99-03 PMIDs form the original same-trial report group; 148 participants, durations and endpoints are not added as new independent populations. The five-RCT/1397-participant review's inclusion version, contributions and overlap remain unread and are not counted independently alongside original trials. | Material | Public availability | Access status | Actually read | Notes | |---|---|---|---|---| | Stored summary | Local preserved copy | Local preserved summary acquired | Summary read | Relevant evidence, bibliography and Korean and English context from the complete specified preserved summary were linked for production. External full texts were not read. | | External abstract | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Only original bibliography links were preserved; current access, reading and link verification were not performed. | | External full text or follow-up report | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Original links connect bibliographic identities and do not indicate direct full-text acquisition or reading. Follow-up of the same trial is not a new independent population. | | External registration | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | The original registration statement that it predates registration requirements and is absent was preserved; current registration existence or absence was not directly queried or certified. | | Required materials: protocol or review plan | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Required scope: failure and survival definitions, flow and censoring, interim plans and adjudication, exposure and targets, safety and Amgen roles; original-material existence, public availability and acquisition remain unverified. | | Required materials: SAP, interim or synthesis analyses | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Endpoint hierarchy and models, interim analysis, censoring and missingness, or review coding, inclusion and synthesis methods remain unread. | | Required materials: results, denominators, safety and regulation | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Direct original endpoint- and timepoint-specific results, flow, safety and relevant regulatory materials are needed; FDA links or contents were not newly queried or created. | | Required materials: funding and COI | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Report-specific sponsorship, free drug, operational, analysis and manuscript roles and review declarations were not directly read. | ### R02 Henke M, Laszig R, Rübe C, et al. Erythropoietin to treat head and neck cancer patients with anaemia undergoing radiotherapy. Lancet. 2003;362:1255-1260. PMID: 14575968. [PubMed](https://pubmed.ncbi.nlm.nih.gov/14575968/) Materials actually read: Summary read; external abstracts and full texts not read Findings recorded in the summary: Original notation: 351 (180/171), ITT. Locoregional PFS adjusted RR 1.62 (95% CI 1.22–2.14), P=.0008; survival RR 1.39 (1.05–1.84), in the original harmful direction. Details of models, exposure and intervals remain unread. Record level: Not recorded Independent replication: DAHANCA's same population, failure, survival, Hb, safety, interim and follow-up context; original independence statements are historical and do not currently certify unique families. ENHANCE's original drug, population, locoregional PFS and survival context is not added to other trial values or denominators. The two initial and long-term RTOG 99-03 PMIDs form the original same-trial report group; 148 participants, durations and endpoints are not added as new independent populations. The five-RCT/1397-participant review's inclusion version, contributions and overlap remain unread and are not counted independently alongside original trials. | Material | Public availability | Access status | Actually read | Notes | |---|---|---|---|---| | Stored summary | Local preserved copy | Local preserved summary acquired | Summary read | Relevant evidence, bibliography and Korean and English context from the complete specified preserved summary were linked for production. External full texts were not read. | | External abstract | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Only original bibliography links were preserved; current access, reading and link verification were not performed. | | External full text or follow-up report | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Original links connect bibliographic identities and do not indicate direct full-text acquisition or reading. Follow-up of the same trial is not a new independent population. | | External registration | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | The original registration statement that it predates registration requirements and is absent was preserved; current registration existence or absence was not directly queried or certified. | | Required materials: protocol or review plan | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Required scope: PFS and survival definitions, actual denominators and ITT, adjustment, exposure and targets, safety and Roche roles; original-material existence, public availability and acquisition remain unverified. | | Required materials: SAP, interim or synthesis analyses | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Endpoint hierarchy and models, interim analysis, censoring and missingness, or review coding, inclusion and synthesis methods remain unread. | | Required materials: results, denominators, safety and regulation | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Direct original endpoint- and timepoint-specific results, flow, safety and relevant regulatory materials are needed; FDA links or contents were not newly queried or created. | | Required materials: funding and COI | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Report-specific sponsorship, free drug, operational, analysis and manuscript roles and review declarations were not directly read. | ### R03 Machtay M, Pajak TF, Suntharalingam M, et al. Radiotherapy with or without erythropoietin for anemic patients with head-and-neck cancer. Int J Radiat Oncol Biol Phys. 2007. PMID: 17716826; long-term follow-up PMID: 25670542. [PubMed](https://pubmed.ncbi.nlm.nih.gov/25670542/) Materials actually read: Summary read; external abstracts and full texts not read Findings recorded in the summary: 148 randomized. Three-year failure 44% versus 36% (P=.56), five-year failure 46.2% versus 39.4% (P=.42), OS 36.9% versus 38.2% (P=.54), originally described as no benefit. Initial PMID 17716826 and long-term PMID 25670542 concern the same trial and are not added as new populations. Record level: Not recorded Independent replication: DAHANCA's same population, failure, survival, Hb, safety, interim and follow-up context; original independence statements are historical and do not currently certify unique families. ENHANCE's original drug, population, locoregional PFS and survival context is not added to other trial values or denominators. The two initial and long-term RTOG 99-03 PMIDs form the original same-trial report group; 148 participants, durations and endpoints are not added as new independent populations. The five-RCT/1397-participant review's inclusion version, contributions and overlap remain unread and are not counted independently alongside original trials. | Material | Public availability | Access status | Actually read | Notes | |---|---|---|---|---| | Stored summary | Local preserved copy | Local preserved summary acquired | Summary read | Relevant evidence, bibliography and Korean and English context from the complete specified preserved summary were linked for production. External full texts were not read. | | External abstract | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Only original bibliography links were preserved; current access, reading and link verification were not performed. | | External full text or follow-up report | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Original links connect bibliographic identities and do not indicate direct full-text acquisition or reading. Follow-up of the same trial is not a new independent population. | | External registration | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Only original string NCT00004917 was preserved; registration contents, timing and plans remain unread. | | Required materials: protocol or review plan | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Required scope: initial and follow-up flow, endpoint- and duration-specific denominators and CIs, exposure and analyses and NCI/Ortho roles; original-material existence, public availability and acquisition remain unverified. | | Required materials: SAP, interim or synthesis analyses | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Endpoint hierarchy and models, interim analysis, censoring and missingness, or review coding, inclusion and synthesis methods remain unread. | | Required materials: results, denominators, safety and regulation | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Direct original endpoint- and timepoint-specific results, flow, safety and relevant regulatory materials are needed; FDA links or contents were not newly queried or created. | | Required materials: funding and COI | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Report-specific sponsorship, free drug, operational, analysis and manuscript roles and review declarations were not directly read. | ### R04 Bohlius J, Schmidlin K, Brillant C, et al. Erythropoietin or darbepoetin for patients with cancer. Cochrane review subgroup of head and neck radiotherapy trials. PMID: 19588382. [PubMed](https://pubmed.ncbi.nlm.nih.gov/19588382/) Materials actually read: Summary read; external abstracts and full texts not read Findings recorded in the summary: Original review size: five RCTs/1,397 participants. OS Peto OR 0.73 (95% CI 0.58–0.91), locoregional PFS OR 0.63 (0.49–0.80), originally favoring radiotherapy alone. Inclusion, contributions, timing and independent-population counts remain unverified. Record level: Not recorded Independent replication: DAHANCA's same population, failure, survival, Hb, safety, interim and follow-up context; original independence statements are historical and do not currently certify unique families. ENHANCE's original drug, population, locoregional PFS and survival context is not added to other trial values or denominators. The two initial and long-term RTOG 99-03 PMIDs form the original same-trial report group; 148 participants, durations and endpoints are not added as new independent populations. The five-RCT/1397-participant review's inclusion version, contributions and overlap remain unread and are not counted independently alongside original trials. | Material | Public availability | Access status | Actually read | Notes | |---|---|---|---|---| | Stored summary | Local preserved copy | Local preserved summary acquired | Summary read | Relevant evidence, bibliography and Korean and English context from the complete specified preserved summary were linked for production. External full texts were not read. | | External abstract | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Only original bibliography links were preserved; current access, reading and link verification were not performed. | | External full text or follow-up report | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Original links connect bibliographic identities and do not indicate direct full-text acquisition or reading. Follow-up of the same trial is not a new independent population. | | External registration | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | The original registration value not applicable was preserved; current registration existence or absence was not directly queried or certified. | | Required materials: protocol or review plan | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Required scope: version, inclusion and contributions, unique families, OR coding, synthesis, heterogeneity, sponsorship and COI; original-material existence, public availability and acquisition remain unverified. | | Required materials: SAP, interim or synthesis analyses | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Endpoint hierarchy and models, interim analysis, censoring and missingness, or review coding, inclusion and synthesis methods remain unread. | | Required materials: results, denominators, safety and regulation | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Direct original endpoint- and timepoint-specific results, flow, safety and relevant regulatory materials are needed; FDA links or contents were not newly queried or created. | | Required materials: funding and COI | Current public availability and detailed existence remain unverified | Access not attempted; not obtained | Not read | Report-specific sponsorship, free drug, operational, analysis and manuscript roles and review declarations were not directly read. | ## Review criteria and methods These are Chamgap's own application conditions, informed by the relevant sections of official methodological guidance. They do not constitute official assessment-tool certification or a new efficacy score. - M03 [Cochrane Handbook 6.5 (2024), chapter updated 2023-08](https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current/chapter-03) · §3.1, §3.2.1–3.2.3, MECIR C5/C7/C13 · Clearly record the question, eligibility conditions and reasons for changes, and distinguish variations in interventions. · Methodological principles stated in existing pilot materials and condition templates were referenced. The methodological text was not reviewed again in this work. - M04 [Cochrane Handbook 6.5.1 (2025), chapter updated 2025-03](https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current/chapter-04) · §4.5, §4.6.1–4.6.3, MECIR C42 · Distinguish studies from reports and link them; do not omit a study solely because results are unreported. · Methodological principles stated in existing pilot materials and condition templates were referenced. The methodological text was not reviewed again in this work. - M08 [Cochrane Handbook 6.5 (2024), chapter updated 2019-10](https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current/chapter-08) · §8.1–8.2.3, Table 8.2.a · Define the outcome and effect of interest, then examine randomization, deviations, missing data, measurement and selective reporting. · Methodological principles stated in existing pilot materials and condition templates were referenced. The methodological text was not reviewed again in this work. - M15 [Cochrane Handbook 6.5 (2024), chapter updated 2023-08](https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current/chapter-15) · §15.2.4, §15.3.1–15.3.3 · Distinguish effect estimates, intervals, importance and applicability. Sample size alone does not determine precision. · Methodological principles stated in existing pilot materials and condition templates were referenced. The methodological text was not reviewed again in this work. - M19 [Cochrane Handbook 6.5 (2024), chapter updated 2019-10](https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current/chapter-19) · §19.1.2, §19.2.1–19.2.4 · Harm questions may require different designs, populations and follow-up from efficacy questions. Case reports do not establish incidence rates. · Methodological principles stated in existing pilot materials and condition templates were referenced. The methodological text was not reviewed again in this work. ## What the statuses mean - Met: The original content has been reviewed as meeting the conditions. This status is not used in this preparation. - Partly verified: Some information in the stored summary was linked during preparation; clinical verification or review is not complete. - Unverified: Required direct values, definitions or materials have not been established within the current storage and reading scope. This does not mean zero effect, no papers or failed access. - Review incomplete: Review of existing content or actual review incomplete. - Not reported: Used only when original materials were actually read and the outcome was judged unreported. Not used for external materials unread in this work. - Inaccessible: Used only when an actual access attempt has a recorded failure. Not used for access that was not attempted in this work. - Insufficient evidence for the conditions: Indicates insufficient direct materials for the relevant scope or conditions; it is not a new grade. - Not applicable: Used when the reason it does not apply to the original question is stated. - Conflicting data: Retains the scope of conflicting statements in original materials; it is not an arbitrary reassessment. ## Change history 2026-10-11 · Public page/materials 1.0. A public view and materials were prepared from the existing 44 items. No new clinical search, verification-status change or scoring was performed.