Differences in Clinicopathologic Features and Response to Neoadjuvant Radiotherapy for Early- Versus Late-Onset Stage II-III Rectal Adenocarcinoma

Abstract

The incidence of rectal cancer in adolescents and young adults (AYA) is increasing. While randomized trials in older adults have established neoadjuvant radiotherapy (nRT) as a standard treatment for stage II-III disease, there is limited evidence supporting this regimen in AYA. We hypothesized that AYA with rectal cancer have a different underlying biology and response to nRT compared to older adults. To investigate this, we conducted a database study comparing clinicopathologic features at diagnosis and response to treatment in AYA vs older adults with stage II-III rectal adenocarcinoma. Patients diagnosed with stage II-III rectal adenocarcinoma between 2004-2016 were identified in the National Cancer Database. Early-onset (EO-RC) and late-onset rectal cancer (LO-RC) were defined as aged 15-39 and 40-90 at diagnosis. For overall survival (OS) analyses, the LO-RC cohort was subdivided into middle-onset (age 40-65) and late-onset (age 66-90) groups to adjust for differences in life expectancy and selection bias. Chi-squared and t-tests were used to compare clinicopathologic features between groups. Unadjusted OS was estimated using the Kaplan-Meier method, with groups compared by log-rank test. Of 55,093 patients with stage II-III rectal adenocarcinoma, 2,785 (5%) had EO-RC. Patients with EO-RC had fewer comorbidities (7% vs 21% Charlson/Deyo score ≥1, p<0.001) and presented with more advanced disease at diagnosis, including a higher rate of nodal metastases (53% vs 41%, p<0.001). Among patients who received nRT, those with EO-RC had similar rates of pathologic complete response (11% vs 12%, p = 0.17) and sphincter-preserving resection (71% vs 71%, p = 0.56) compared to LO-RC, but were slightly less likely to have pathologic downstaging of the primary tumor (47% vs 50%, p = 0.04) and nodal metastases (60% vs 66%, p<0.001). The EO-RC cohort had better 90-day post-operative mortality (0.3% vs 1.6%, p<0.001) than the LO-RC cohort. Unadjusted OS analyses showed that nRT receipt was not associated with OS in early-onset or middle-onset patients (p>0.05); however, nRT was associated with improved OS in LO-RC patients (p<0.001). The estimated 10-year OS for patients aged 15-39, 40-65, and 66-90 at diagnosis was 67%, 60%, and 37%, respectively (p<0.001). AYA with stage II-III rectal adenocarcinoma were more likely to present with nodal metastases, slightly less likely to have pathologic downstaging following nRT, and demonstrated no OS benefit from nRT as compared to older adults. Since AYA have fewer medical comorbidities and higher long-term survival, they may be at increased risk of late radiation-related toxicities. Further research to appropriately identify a cost-effective screening strategy in high-risk AYA is crucial to decrease morbidity and mortality.