Molecular therapy selection in treatment-refractory advanced cancers: A retrospective cohort study determining the utility of TOPOGRAPH knowledge base.

Abstract

3073 Background: Comprehensive genomic profiling (CGP) is increasingly used to guide therapy selection in advanced cancer patients who have exhausted standard therapy options. Here we assess the utility of Therapy Oriented Precision Oncology Guidelines for Recommending Anticancer Pharmaceuticals (TOPOGRAPH) to guide matching of drug treatments based on CGP in this setting. Methods: This study was conducted in an Australia-wide precision oncology program, the Molecular Screening and Therapeutics Study (MoST, ANZCTR registration ACTRN12616000908437). All patients with advanced cancer after exhausting standard treatments underwent CGP in 2016-2021 were stratified into cohort A (no further therapy received) and B (received ≥1 therapy after CGP). The primary outcome was overall survival (OS) estimated by the Kaplan-Meier method, using the log rank test to assess between-group differences. TOPOGRAPH matched the treatment history to the CGP results, stratified into clinically active (Tiers 1-3, T1-3), investigational (T3B/4), inactive (R2) or unmatched groups. Results: Over a median follow-up of 21.7 months (mo) for 2852 patients (75% with rare cancers, n = 2150), the median OS (mOS) from the date of CGP result was 7.0 mo (95% CI 6.4-7.6) for cohort A (n = 1562) and 15.8 mo (95% CI 14.5-16.9) for cohort B (n = 1290). In both cohorts, patients with CGP results matching any TOPOGRAPH tier (T1-4) had shorter OS compared to patients without a matching tier (A: 6.4 v 20.5 mo, hazard ratio for death [HR] 2.15, p<0.001; B: 14.7 v 23.6 mo, HR 1.43, p<0.001). Patients in cohort B receiving matched therapy (n=342, 27%) had a longer mOS than 948 patients who received only unmatched therapy (16.9 v 10.4 mo, HR 0.70, p<0.001). For CGP results matched to T1-3, 122 patients who received a T1-3 therapy had a significantly longer mOS than those who received unmatched therapy (22.1 v 9.8 mo, HR 0.51, p<0.001). For CGP results matched only to T3B/4, a trend toward longer mOS was observed in patients receiving matched therapy in T3B/4 (n = 138, 14.5 v unmatched 10.0 mo, HR 0.81, P = 0.07). In tier-matched analysis, the mOS were not significantly different between patients who received genomics matched v unmatched therapy in T3B (matching outside cognate histotypes, n = 48 v 508, 11.9 v 9.7 mo, HR 0.84, p = 0.36) and T4 (n = 32 v 134; therapy with preclinical/early clinical evidence, 17.1 v 12.2 mo, HR 0.69, p = 0.17). Conclusions: TOPOGRAPH is prognostic and likely predictive of treatment effect based on CGP, supporting its utility in guiding molecular therapy selection in patients who have exhausted standard treatment options. [Table: see text]