This secondary analysis examined correlations between baseline tumor burden and outcomes in a phase 2 basket trial cohort of 722 patients with rare/ultrarare malignancies and measurable disease treated with nivolumab plus ipilimumab. The study was conducted across over 1000 sites through the National Cancer Institute/Southwest Oncology Group. No comparator group was reported for this analysis.
In multivariable analysis, larger baseline tumor burden (≥12.9 cm vs. 1.0-4.8 cm) correlated with shorter overall survival (hazard ratio 1.64, 95% CI 1.02-1.72, p=.032). No correlation was found with progression-free survival. Higher baseline tumor burden quartiles showed only a weak negative association with any tumor regression at first scan (Fisher exact test p=.09), and no significant interaction was found between tumor burden and tumor regression (p for interaction >.65).
Safety and tolerability data were not reported for this analysis. Key limitations include its observational nature within a trial, the absence of testing for an intervention on tumor burden, and specificity to rare cancer types treated with dual immune checkpoint inhibitor therapy. The analysis does not establish causation.
Practice relevance is limited by the phase 2 design, secondary analysis approach, and cohort specificity. The finding suggests baseline tumor burden may be a prognostic factor for overall survival in this particular context, but it does not inform treatment selection or predict early response to therapy.
View Original Abstract ↓
BACKGROUND: It has been suggested that baseline tumor burden may correlate with immune checkpoint inhibitor (ICI) outcome for individual tumor types in which ICIs are standardly used. The authors investigated whether pretreatment tumor burden correlates with overall survival (OS), progression-free survival (PFS), and tumor regression among patients who had rare cancers treated with dual ICIs.
METHODS: Southwest Oncology Group study S1609 was a phase 2, National Cancer Institute/Southwest Oncology Group basket study (>1000 sites) evaluating nivolumab plus ipilimumab in 53 cohorts of patients who had rare/ultrarare malignancies (ClinicalTrials.gov identifier NCT02834013). Overall, 722 patients were included in this secondary analysis, all of whom had measurable disease (Response Evaluation Criteria in Solid Tumors, version 1.1). Baseline tumor burden, defined as the sum of the greatest dimensions of target lesions at study registration, was analyzed based on quartiles observed in the data. The number of target lesions was also considered a secondary tumor burden measure. End points included OS and PFS.
RESULTS: Larger baseline tumor burden correlated with shorter OS, but not PFS (multivariable analysis). Higher baseline tumor burden quartiles had only a weak negative association with any tumor regression at first scan (Fisher exact test, p = .09), and multivariable analyses further indicated that both tumor burden and any tumor regression at first posttreatment scan were independently associated with OS in multivariable analysis (comparing a baseline tumor size ≥12.9 cm vs. 1.0-4.8 cm; hazard ratio, 1.64; 95% confidence interval, 1.02-1.72; p = .032), but there was no evidence of an interaction between tumor burden and any tumor regression at the first scan (p for interaction > .65).
CONCLUSIONS: Larger baseline tumor burden was associated with worse OS, but not PFS, and was not predictive of tumor regression after dual ICI therapy in a large cohort with rare cancer types.