SS202 - Response to Neoadjuvant Therapy Predicts Metastatic Burden and Survival After Distant Metastasis in Locally Advanced Rectal Cancer

Purpose/Background: The grade of tumor response to neoadjuvant therapy (NAT) has been associated with different rates of local recurrence (LR) and distant metastasis (DM). Survival after DM for patients with different grades of tumor response at the primary tumor site has not been previously investigated. Here we report overall survival after DM in patients with locally advanced rectal cancer treated with NAT and total mesorectal excision or watch and wait (WW) based on tumor response.

Methods/Interventions: Patients with LARC treated with NA chemotherapy, chemoradiation (CRT) or total neoadjuvant therapy (TNT) who subsequently developed DM were identified. Based on treatment response at the primary site, patients were classified as: (1) complete response (CR, clinical or pathological), (2) incomplete clinical response (iCR), or (3) regrowth during WW. Baseline clinical, pathologic and metastatic features were analyzed. Overall survival (OS) was measured from the time of first DM to death from any cause or last follow-up. Univariable Cox models assessed OS association with (1) baseline tumor characteristics including tumor grade, pT stage, nodal positivity, cEMVI, (2) outcome-related variables at the time of DM, including DM site, LR by DM, and regrowth by DM, and (3) time varying covariates relative to DM including post-DM regrowth and LR. Additionally, the associations of tumor response with baseline tumor characteristics and DM site were assessed with Fisher's exact test.

Results/Outcomes: A total of 389 patients developed DM (52 CR, 280 iCR, 57 regrowth). Compared with iCR, regrowth patients more often had lower pT stage (pT0-1 18% vs 5%, p = 0.017) and were more frequently node-negative (64% vs 42%, p < 0.001), while margin status was similar between groups (p=0.800). Lung (69%) and liver (44%) were the most frequent sites of DM in all groups. A single metastatic site was more common in CR (65%) and regrowth (68%) than iCR (47%, p=0.002). Median time to DM from restaging was 1.0 years (range: 0-7.7) for CR, 0.9 (0-9.4) for iCR and 1.2 (0.1-6.9) for regrowth. LR occurred in 61 (22%) iCR and 8 (14%) regrowth patients over the study period. Median OS for the whole cohort was 3.6 years (95% CI: 3.2, 4.3) with a median follow-up of 3.1 years among survivors. On univariable Cox analysis, compared to iCR patients, OS did not differ for CR patients (HR: 1.0, [95% CI: 0.5, 2.1], p =0.600) and for regrowth patients (HR: 1.1, [95% CI: 0.7, 1.7], p = 0.800). Among patients who have an undefined treatment response at DM date, OS was worse for those who later developed regrowth (HR: 3.70, [95% CI: 1.5, 9.0], p = 0.004) vs sustained CR. cEMVI (HR 1.5, [95% CI: 1.1, 2.2], p=0.015), and LR, whether present at DM (HR 2.0, [95% CI: 1.3, 3.1], p=0.001) or occurring after the diagnosis of DM (HR 2.1, [95% CI: 1.4, 3.3], p< 0.001), were significantly associated with worse OS, while low vs high tumor grade (HR 0.41, [95% CI: 0.3, 0.6], p< 0.001) was associated with improved OS.

Conclusion/Discussion: Patients achieving a CR after NAT were more likely to present with oligometastatic disease and demonstrated improved survival after DM compared to patients with incomplete response or tumor regrowth during WW. Survival after the diagnosis of metastasis was lower among patients with local recurrence. These findings suggest that primary tumor response to NAT reflects underlying tumor biology and influences prognosis even after metastasis. Incorporating primary tumor response into risk stratification may improve prognostic assessment and guide surveillance strategies.