29P - Considering intra-patient response variability in clinical trials: Implications for treatment efficacy and survival

Background

Clinical trials often use Response Evaluation Criteria in Solid Tumors (RECIST) to assess treatment effectiveness. RECIST measures the sum of predefined target lesion diameters on radiologic imaging longitudinally and categorizes patients based on the sum of individual lesion changes. For this study, we leveraged a unique dataset in soft tissue sarcoma (STS), a model of tumor heterogeneity, to investigate the potential relevance of variability in intra-patient lesion responses.

Methods

Lesions in advanced STS patients were assessed according to RECIST before and after 2 cycles of chemotherapy for patients with 2 or more target lesions. Lesions with a diameter increase of 20% or more were labeled ‘non-responders' and the rest were ‘responders'. Patients were categorized as ‘homogeneous' if all lesions responded or ‘heterogeneous' if at least one did not. Survival differences between patients were analyzed using the log-rank test.

Results

Lesion-specific treatment responses were assessed in 466 patients with advanced STS enrolled in an open-label clinical trial (NCT01440088). Among patients who, according to RECIST, had “stable” disease (SD), overall survival (OS) was lower in the heterogeneous versus homogeneous cohorts (p=0.01). A majority of patients (78.7%) had all lesions responding to therapy, whereas the remaining patients (oligoprogressors) had 1 (18.8%) or 2 (2.5%) non-responding lesions.

Conclusions

In advanced STSs, oligoprogression correlated with poorer survival outcomes compared to patients without progressive lesions, despite both groups being classified as “stable” according to RECIST. Such oligoprogressive patients may benefit from local targeted therapy in addition to systemic treatment in order to “convert” them to non-progressors. Our results indicate that a lesion-specific approach could potentially alter the outcomes of clinical trials, suggesting a need for a more nuanced evaluation method of treatment response.

Clinical trial identification

NCT01440088 (Our retrospective study uses the data from this trial.)

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

National Cancer Institute of the National Institutes of Health under Award Number P50CA272170.

Disclosure

All authors have declared no conflicts of interest.