Cohort studycited by the Tayebat archivefull text checked★ key study A-CHM-07 in-depth review
Only a small share of second cancers in adults is caused by radiotherapy
Proportion of second cancers attributable to radiotherapy treatment in adults: a cohort study in the US SEER cancer registries.
Overview
Berrington de Gonzalez A et al.. Lancet Oncol 2011 · PMID 21454129 · DOI 10.1016/S1470-2045(11)70061-4
About 5 extra cancers per 1,000 treated patients within 15 years; 8% of second cancers in irradiated patients.
Does the study support the specific claim? Partly · Methodological quality: moderate
Plain-language summary
Study explained simply
- What did the researchers want to find out?
Estimate radiotherapy-attributable second cancers.
- How was the study done?
SEER cohort.
- How many people, animals or samples were studied?
647,672 survivors.
- What was actually found?
8% attributable; 5/1000 by 15 years.
- How large was the observed effect?
RR 1.08-1.43.
- What do the results mean in everyday life?
The risk is real but small compared with the benefit of cure.
- What are the study's limitations?
No randomisation; no data on smoking, chemotherapy or doses; treatments before 2003.
- What does it say about Elawady's specific claim?
Limits the 'heavy price' framing.
- Do the results apply to Germany and Europe?
High.
- What can explicitly not be concluded from this study?
Risk for an individual patient.
Scientific in-depth review
1. Full reference
- Authors
- Berrington de Gonzalez A, Curtis RE, Kry SF, Gilbert E, Lamart S, Berg CD, Stovall M, Ron E
- Journal
- Lancet Oncol
- Year
- 2011
- Identifier
- PMID 21454129 DOI 10.1016/S1470-2045(11)70061-4 · PMC3086738
- Verification
- verified – PubMed E-utilities efetch, 2026-10-25
- Retracted
- no
- Corrections
- none
- Extent checked
- full text
- Conflicts of interest and funding
Funding: intramural research program of the US National Cancer Institute; sponsors had no role in design, analysis, interpretation, writing or submission. Conflicts of interest: the authors declared none.
2. Research question
Proportion of second solid cancers attributable to radiotherapy in adults.
3. Study design and population
- Study design
- Cohort study – Registry-based cohort (SEER) of 5-year survivors, radiotherapy vs no radiotherapy
- Population
Adults aged 20+ with a first primary invasive solid cancer at 15 sites routinely treated with radiotherapy (oral/pharynx, salivary gland, rectum, anus, larynx, lung, soft tissue, breast, cervix, endometrium, prostate, testis [seminoma], eye/orbit, brain/CNS, thyroid), SEER-9 registries, surviving >=5 years
- Country
USA
- Period
Diagnosis 1973-2002; follow-up for second cancers to 31 Dec 2007 (1978-2007); mean follow-up 12 years from diagnosis (SD 4.5, range 5-24), i.e. about 7 years beyond the 5-year landmark
- Sample size
647,672 5-year survivors
- Inclusion and exclusion criteria
Sites with >20% of patients receiving radiotherapy in the first course and >200 second cancers. Excluded: survival <5 years, missing radiotherapy or stage data (except prostate, brain, lung), haematological first and second cancers, non-seminoma testicular and small-cell lung cancers (chemotherapy confounding), extremity soft-tissue sarcoma; censoring at age 85.
4. Methods and endpoints
- Exposure / intervention
Radiotherapy (external beam, brachytherapy or combined) as part of the initial treatment course, as coded by SEER (yes/no); no individual dose data; second cancer sites grouped by estimated dose (>5 Gy, 1-5 Gy, <1 Gy) from standard protocols for nine first-cancer sites
- Comparator
No radiotherapy
- Endpoints
Second primary solid cancer (all sites combined; also by dose group of the second-cancer site)
- Statistical methods
Poisson regression (EPICURE AMFIT) with general-population expected numbers as offset (adjusting sex, attained age, calendar period), stratified by stage, age at and year of diagnosis; effect modification by age, latency, calendar year; sensitivity analysis restricted to surgically treated patients; excess cancers and attributable risk derived from model, expressed relative to second cancers in all 1+ year survivors (assumes zero radiation-related cancers in the first 5 years).
5. Actual results
- Results
60,271 (9%) of 647,672 five-year survivors developed a second solid cancer. RR for radiotherapy vs none >1 for every site, from 1.08 (95% CI 0.79-1.46) eye/orbit to 1.43 (1.13-1.84) testis (seminoma). RR for high-dose (>5 Gy) second-cancer sites from 1.03 (0.71-1.47) after thyroid to 1.78 (1.42-2.22) after cervical cancer; significant dose trend for breast, endometrial and prostate (p<0.0001). RRs fell with older age at diagnosis and rose with latency (e.g. endometrium 1.30 at 5-9 y to 2.18 at 15+ y). 3,266 (2,862-3,670) excess second solid cancers = 8% (7%-9%) of 42,294 second solid cancers in irradiated 1+ year survivors; range 4% (-11% to 20%) eye/orbit to 24% (9%-37%) testis; 17% (10%-23%) after cervical cancer. 3 excess cancers/1,000 treated by 10 years, 5/1,000 by 15 years. 54% of excess cancers in high-dose sites; excluding the low-dose group lowers the attributable risk (nine sites) from 7% (6%-8%) to 5% (4%-6%). Over half of excess cancers (55%, n=1,791) in breast and prostate survivors.
- Absolute risks
3 excess second cancers per 1,000 patients treated with radiotherapy by 10 years and 5 per 1,000 by 15 years after diagnosis (data not shown in tables); 3,266 excess vs 42,294 observed second solid cancers in irradiated patients
- Authors' conclusion
A relatively small proportion of second cancers (<10%) in adult survivors are likely related to radiotherapy, suggesting most are due to other factors such as lifestyle or genetics; absolute risks are typically a few excess cancers per 1,000 treated and small compared with treatment benefits; second-cancer risks of newer techniques such as IMRT still need study.
6. Strengths
- Very large population-based registry (647,672 five-year survivors), 15 first-cancer sites analysed systematically
- Internal comparison (radiotherapy vs no radiotherapy) adjusted for stage, age, year of diagnosis
- Exclusion of the first 5 years to reduce surveillance bias
- Plausibility checks: dose-group gradients, age-at-exposure and latency patterns
- Sensitivity analysis in surgically treated patients changed RRs by <=10% (except eye/orbit)
- Reports attributable fractions and cumulative excess absolute risk
7. Weaknesses
- No randomisation; confounding by indication (authors)
- No data on smoking, chemotherapy or hormonal therapy; higher RRs for smoking-related second cancers suggest residual confounding by smoking (authors)
- Only initial-course radiotherapy recorded; under-ascertainment possible; no individual doses, dose groups crude and likely misclassified
- Treatment before 2003; IMRT and modern techniques not evaluated
- Attributable fraction assumes no radiation-related cancers in first 5 years and is expressed against 1+ year survivors
- Cumulative absolute risk figures given as 'data not shown'
- Many subgroup comparisons; time variables correlated (authors urge caution)
8. Risk of bias
- Orienting tool
- ROBINS-E
- Type of assessment
- structured assessment based on the full text by a single reviewer (no formal double review)
- Full formal assessment
- no
- Overall risk of bias
- some concerns
- Rationale
Unchanged overall judgement. Residual confounding could bias in either direction; absolute excess estimates are model-dependent. Single-reviewer structured judgement from the full text; a formal assessment requires two independent reviewers and a protocol.
| Domain | Judgement | Rationale |
|---|---|---|
| Confounding | some concerns | Adjusted for stage, age, year of diagnosis, attained age and period; no smoking, chemotherapy or hormone data. Smoking-related second cancers show higher RRs (possible confounding). Direction of residual confounding may vary by site (authors). Similar attributable risk for breast cancer from pooled randomised trials (8%) argues against strong confounding. |
| Measurement of exposure | some concerns | SEER initial-course radiotherapy coding (yes/no) is known to be incomplete; no doses; dose groups assigned from standard protocols. |
| Selection of participants | low | Population-based registry; 5-year landmark applied equally; exclusions prespecified and reasoned. |
| Post-exposure interventions | some concerns | Subsequent treatments (chemotherapy, hormones) not recorded and may differ by radiotherapy status. |
| Missing data | some concerns | Patients with missing radiotherapy or stage excluded (except prostate, brain, lung, where stage often unknown, e.g. 44% unknown for one site); extent of exclusions not quantified in the text. |
| Measurement of outcome | low | Registry-ascertained second primary cancers with stable coding rules; censoring at 85 for under-ascertainment; surveillance bias reduced by 5-year exclusion. |
| Selection of reported result | low | Systematic analysis of all eligible sites with full tables; some figures (cumulative excess by 10/15 years, dose-group excess) 'data not shown'. |
Full-text review (25 October 2026): Added detailed inclusion/exclusion rules, statistical method, site-specific attributable fractions (4%-24%), high-dose site RRs and dose/latency trends, 3/1,000 excess at 10 years, smoking-confounding signal and surgery-restricted sensitivity analysis; clarified follow-up (mean 12 years from diagnosis; abstract's '7 additional years'); corrected denominator 42,294 (table; 42,302 in text); full funding/COI. Bias overall unchanged (ROBINS-E some concerns), now domain-level.
9. Assessment of the statistical interpretation
Relative risks are modest (1.08-1.43 overall; up to 1.78 at high-dose sites) and the attributable fraction is 8% overall, i.e. about 92% of second cancers in irradiated adults are not attributed to radiotherapy. Absolute excess is about 5 per 1,000 treated by 15 years. Attributable fractions are higher where patients are young and fields large (e.g. testis 24%, cervix 17%). Wide CIs for small sites (eye/orbit, anus, salivary gland).
10. Relevance to the specific claim
Quantifies second-cancer risk as real but modest in absolute terms.
11. Independent contrary findings
Itself limits the archive framing.
12. Transferability to Germany and Europe
High.
13. Overall methodological judgement
Large, well-conducted registry analysis providing an absolute-risk anchor; residual confounding acknowledged. The archive cites it for a 'heavy price' of second cancers, whereas the authors conclude the radiotherapy-related share is small relative to benefits.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | moderate |
| Risk of bias | some concerns |
| Precision of results | high |
| Directness | direct |
| Reproducibility | partly replicated |
| Consistency with the overall evidence | consistent |
| Regional transferability | high |
| Adequacy of the interpretation (by the archive) | partly correct but over-interpreted |
14. Open questions
Second-cancer risks after IMRT, proton and other post-2002 techniques; risks for patients irradiated at young ages; dose-based individual estimates.
15. Linked fact checks
- Chemotherapy and radiotherapy can cause second cancers (e.g., breast cancer after chest radiation for Hodgkin disease, therapy-related myeloid leukaemia) years after cure. – Studies with mixed or unclear results
- The archive's 19 chemo/radiation studies are presented on the homepage as 'scientific studies confirming what Dr. Diaa proposed'; the harm-only study selection implies that the harms of chemotherapy and radiotherapy outweigh their benefits. – Studies contradicting or limiting the claim
Use by the Tayebat archive
Note: according to the archive, the studies were compiled by the archive operators. Whether Dr. Elawady himself cited this study is not established.
- What the archive says about the study
Listed in the bibliography (secondary cancers as part of a 'heavy price').
- A. What was actually studied?
Share of second cancers due to radiotherapy.
- B. What did the authors find?
8% of second cancers in irradiated patients; 5 per 1,000 by 15 years.
- C. What conclusions do the authors themselves draw?
Relatively small proportion; most second cancers due to lifestyle or genetics.
- D. What does the archive derive from it?
Supports the narrative of a heavy secondary-cancer burden.
- E. Is this derivation scientifically justified?
Partly: the study documents a real excess (5 per 1,000 treated by 15 years; up to 24% attributable fraction after testicular seminoma), but the authors conclude that <10% of second cancers relate to radiotherapy and that risks are small compared with treatment benefits; the archive does not report this.
- Adequacy of the interpretation (by the archive)
- partly correct but over-interpreted