Highlights
– In 25,723 five‑year childhood cancer survivors (median follow‑up 28.5 years), colorectal subsequent malignant neoplasms (SMNs) were strongly associated with mean colorectum dose (MCD) and irradiated colorectal volume in a dose‑dependent manner.
– Specific chemotherapy exposures showed independent and dose‑dependent risks: procarbazine demonstrated a steep linear excess rate ratio (ERR) per 1,000 mg/m2, and high cumulative doses of alkylators and platinum were also associated with markedly increased colorectal SMN incidence.
– The study provides colorectum‑specific RT dose/volume thresholds and chemotherapy exposure metrics that can inform contemporary RT planning and stratified long‑term surveillance for higher‑risk survivors.
Background
Long‑term survivors of childhood cancer face elevated risks of second malignant neoplasms (SMNs), which contribute substantially to morbidity and mortality among survivors. Historical therapeutic regimens (1970–1999) combined external beam radiotherapy and cytotoxic chemotherapies that are known carcinogens. Prior childhood cancer survivor studies have reported higher colorectal cancer rates after abdominal or pelvic irradiation, but until now, relationships using colorectum‑specific dosimetric metrics used in modern RT planning (mean dose and dose–volume parameters) have not been reported in a large cohort with detailed chemotherapy dose quantification.
Study design
This report used data from the Childhood Cancer Survivor Study (CCSS), a multi‑institutional retrospective cohort of 5‑year survivors diagnosed between 1970 and 1999. The analytic cohort included 25,723 individuals with a median follow‑up of 28.5 years (range 5.0–48.9). Radiotherapy exposures were reconstructed to estimate colorectum‑specific dosimetry: mean colorectal dose (MCD) and dose–volume metrics (VX Gy = percent colorectum volume receiving ≥X Gy for X = 5, 10, 20, 30, 40 Gy). Chemotherapy exposures were quantified as cumulative doses for procarbazine and platinum agents, cyclophosphamide‑equivalent doses (CED) for alkylating agents, and doxorubicin‑equivalent doses for anthracyclines.
Outcomes were incident colorectal SMNs. Statistical approaches included piecewise‑exponential models to estimate incidence rate ratios (IRRs) and excess rate ratio (ERR) models to examine linear and quadratic dose–response relationships. Reference groups comprised survivors unexposed to the treatment under evaluation.
Key findings
Event burden and cohort: 104 colorectal SMNs occurred among 25,723 survivors across nearly three decades median follow‑up, enabling evaluation of relatively rare events with detailed exposure metrics.
Radiation dose and volume effects
A clear dose–response relationship was observed for MCD. Compared with survivors who received no colorectum irradiation, adjusted IRRs were 3.6 (95% CI, 1.9–6.9) for MCD 10 to <20 Gy and 8.3 (95% CI, 3.9–17.8) for MCD ≥20 Gy. This is a substantial relative risk elevation at clinical dose ranges.
Dose–volume metrics further refined risk attribution. When ≥20% of the colorectum volume was irradiated, IRRs rose with larger irradiated fractions. For the V20 Gy metric, IRRs were 3.8 (95% CI, 1.9–7.6) for 20% to <40% of colorectum irradiated, 4.9 (95% CI, 2.0–12.0) for 40% to <80%, and 8.7 (95% CI, 3.5–21.6) for ≥80% volume irradiated. These volume thresholds provide clinically relevant quantitative targets for dose‑sparing strategies.
Chemotherapy associations
Cumulative chemotherapy exposures were independently associated with colorectal SMN rates, both overall and in analyses restricted to those who did not receive RT. Key adjusted IRRs for highest dose categories included:
- Doxorubicin‑equivalent dose ≥250 mg/m2: IRR 1.8 (95% CI, 1.0–3.0).
- Cyclophosphamide‑equivalent dose (CED) ≥6,000 mg/m2: IRR 3.7 (95% CI, 2.2–6.4).
- Platinum cumulative dose ≥450 mg/m2: IRR 4.5 (95% CI, 2.0–10.1).
- Procarbazine: IRR 6.3 (95% CI, 3.0–13.2) for 4,200–<7,036 mg/m2 and 9.0 (95% CI, 4.3–18.9) for ≥7,036 mg/m2.
Notably, in survivors who did not receive RT, exposure to any platinum agent (IRR 3.8 [95% CI, 1.1–12.7]), any alkylator (IRR 4.8 [95% CI, 1.6–14.4]), or procarbazine (IRR 16.9 [95% CI, 5.9–48.8]) were each associated with large relative increases in colorectal SMN incidence, suggesting that chemotherapy alone (in the absence of RT) can confer meaningful colorectal cancer risk.
Dose–response modeling
ERR modeling indicated linear dose–response relationships for both procarbazine and MCD. The ERR per 1,000 mg/m2 of procarbazine was 73.0% (95% CI, 26.4%–119.6%), and ERR per 1 Gy of MCD was 20.8% (95% CI, 9.0%–32.5%). Quadratic terms did not improve model fit, supporting linearity over the observed exposure ranges.
Clinical magnitude and precision
The reported effect sizes are large and statistically significant for several exposure categories, though confidence intervals are wide in some strata, reflecting the limited number of events (n=104) when stratified by detailed exposure groups. Still, consistent patterns across multiple RT and chemotherapy metrics strengthen causal inference.
Expert commentary: interpretation, strengths, and limitations
Interpretation: This analysis provides actionable, colorectum‑specific dose and volume risk estimates and quantifies chemotherapy dose thresholds associated with markedly increased colorectal SMN risk. The MCD and V20 Gy findings translate directly to modern RT planning: they support strategies that minimize mean colorectum dose and reduce the percentage of colorectum receiving moderate doses (e.g., ≥20 Gy) without compromising tumor control. Chemotherapy findings, especially the steep procarbazine ERR and high risks with large cumulative alkylator or platinum doses, highlight the need for careful selection and long‑term follow‑up when using these agents.
Strengths
- Large, well‑characterized cohort with long median follow‑up and central adjudication of outcomes.
- Individualized, colorectum‑specific dose reconstruction and clinically familiar dose–volume metrics enable translation to modern RT planning systems.
- Detailed cumulative chemotherapy dosing allowed examination of dose thresholds and dose–response relationships.
Limitations
- Historical treatment era (1970–1999) differs from contemporary practice: modern RT techniques (IMRT, volumetric arc therapy, image guidance, and especially proton therapy) and contemporary systemic regimens could change absolute risks even if dose–response relationships remain relevant.
- Observational design can be affected by confounding (by indication, co‑treatments, lifestyle factors, and genetic predisposition); not all potential confounders may have been measured or fully controlled.
- Although the cohort is large, the absolute number of colorectal SMNs (n=104) limits precision in finely stratified analyses and prevents extensive subgroup analyses (e.g., by tumor histology or molecular subtype).
- Information on colorectal screening, surveillance behaviors, and polygenic or monogenic colorectal cancer predisposition was limited and may alter observed risk patterns.
Biological plausibility
Ionizing radiation causes DNA double‑strand breaks and genomic instability, with mutational signatures plausibly contributing to colorectal carcinogenesis. Alkylating agents (and procarbazine specifically) produce DNA adducts and mutations, and platinum agents form crosslinks impairing DNA repair; cumulative exposures increase the likelihood of carcinogenic mutations. The linear dose–response observed for both MCD and procarbazine is consistent with cumulative mutagenic injury driving later malignant transformation.
Clinical implications and recommended actions
For radiation oncologists and pediatric oncology teams:
- When possible, integrate colorectum‑sparing objectives into RT planning. Aim to reduce mean colorectum dose and the percentage of colorectum receiving doses ≥20 Gy, particularly for patients with long expected survivals.
- When high colorectum doses or large irradiated volumes are unavoidable, document colorectum‑specific dose–volume metrics in the treatment record to enable individualized survivorship risk assessment.
For medical oncologists and multidisciplinary teams:
- Recognize that high cumulative procarbazine, alkylator (high CED), and platinum exposures are independently associated with increased colorectal SMN risk. Where clinically appropriate, consider alternative regimens or dose modifications, and discuss long‑term risks with patients and families.
For survivorship clinicians and guideline developers:
- Use the dosimetric and chemotherapy thresholds reported here to refine risk‑stratified colorectal cancer surveillance recommendations. Survivors with MCD ≥10–20 Gy, large V20 Gy percentages, high cumulative procarbazine, platinum, or alkylator exposures probably warrant earlier and more intensive colorectal surveillance than average‑risk populations.
- Existing survivorship frameworks (e.g., Children’s Oncology Group Long‑Term Follow‑Up Guidelines) should consider integrating colorectum‑specific dose/volume and chemotherapy dose thresholds as part of surveillance decision algorithms.
Future research directions
Key priorities include validation of these dose–response relationships in cohorts treated with contemporary RT modalities (including proton therapy), molecular characterization of treatment‑related colorectal tumors to define distinct mutational signatures, integration of host genetic susceptibility and lifestyle risk factors, and evaluation of the effectiveness and cost‑effectiveness of risk‑stratified colorectal surveillance strategies in improving outcomes.
Conclusion
This CCSS analysis supplies clinically actionable, colorectum‑specific RT dose and volume metrics and quantifies chemotherapy exposures associated with significantly increased colorectal SMN risk among childhood cancer survivors. The results support colorectum‑sparing approaches in RT planning and provide evidence to refine survivorship surveillance for high‑risk survivors, while acknowledging the need to interpret absolute risks in light of treatment era differences and individual patient circumstances.
Funding and clinicaltrials.gov
This study used CCSS resources and was conducted by investigators listed in the primary JCO article. Detailed funding sources and acknowledgements are reported in the original publication: Owens CA et al., J Clin Oncol. 2025;43(31):3403–3421. Readers should consult the JCO article for specific grant numbers and participating institution acknowledgements. This was an observational cohort study (CCSS); it is not registered as an interventional clinical trial on ClinicalTrials.gov.
Selected references
1) Owens CA, Ludmir EB, Liu Q, et al. Colorectal‑Specific Radiation Dose and Chemotherapy Risk for Subsequent Colorectal Malignancies in Childhood Cancer Survivors: A Childhood Cancer Survivor Study (CCSS) Report. J Clin Oncol. 2025 Nov;43(31):3403-3421. doi: 10.1200/JCO-25-00531.
2) Oeffinger KC, Mertens AC, Sklar CA, et al. Chronic Health Conditions in Adult Survivors of Childhood Cancer. N Engl J Med. 2006;355:1572–1582. doi:10.1056/NEJMsa060185.
3) Children’s Oncology Group. Long‑Term Follow‑Up Guidelines for Survivors of Childhood, Adolescent, and Young Adult Cancers. Accessed at https://www.children oncology.org (consult most recent version for guideline details).
Author note
This article is a synthesis and critical interpretation of the CCSS report by Owens et al. (J Clin Oncol, 2025) intended for clinicians and survivorship specialists. For full methodological and supplementary details, refer to the original publication.