Can patients with advanced malignancy and poor performance status benefit from nivolumab plus ipilimumab as a palliative treatment?

Abstract

12028 Background: Performance status (e.g. Eastern Cooperative Oncology Group [ECOG] score) is a predictive tool used to determine whether a patient may benefit from cytotoxic chemotherapy. The toxicity profile of immunotherapy is different, and less is known about whether performance status (PS) is similarly associated with toxicity and benefit. Nonetheless, most clinical trials for immune checkpoint inhibitors have excluded patients with poor PS. Emerging data by our group and others has linked poor PS with lack of response and decreased survival with anti-PD-1/L1 agents, but whether combination therapy abrogates the poor outcomes of PD1 only therapy is unknown. Methods: We conducted a retrospective cohort study of patients with metastatic cancer who received ipilimumab plus nivolumab at our institution between January 2014-December 2020. We compared outcomes between those with good PS (Group A, ECOG PS 0-1) and poor PS (Group B, ECOG PS ≥ 2). Our primary outcomes were overall survival (OS) and incidence of grade 3-4 immune-related adverse events (irAEs). Other outcomes included objective response rates and need for hospitalization. We utilized the Kaplan-Meyer method for time to survival analysis and exact Pearson Chi-squared testing for response, toxicity, and hospitalization analysis. Results: A total of 129 patients were identified with a mean age of 59.9 years. Among them were 73 males (57%) and 56 females (43%). Malignant melanoma was the most common malignancy (39%) followed by renal cell carcinoma (27%), small cell lung cancers (10%), non-small cell lung cancers (9%), and the rest with other histologies. 113 patients (87.6%) were in group A and 16 (12.4%) in group B. Across all tumor types, patients in Group B had significantly worse OS compared to group A (HR 0.24 [0.13-0.48], P = < 0.0001). Group B similarly had higher rates of hospitalization compared to group A (94% vs 56%, P = 0.0048). Interestingly, irAEs were not driving these hospitalizations, with a trend toward lower rates of irAEs in the poor PS group (19% vs 46% in Group A, P = 0.057). The overall response rate in group B was numerically lower (6% vs 26% in group A, p = 0.1). Conclusions: Our study showed that using ipilimumab plus nivolumab in patients with poor PS was associated with significantly poorer OS and higher rates of hospitalizations. irAEs were not increased in the poor PS group, suggesting a lack of treatment efficacy seemed to driving these poor outcomes. Extrapolation of clinical trial data for ipilimumab-nivolumab to a broader population including those with poor PS should be done with caution. Unfortunately, our data showed that most patients with poor PS were not adequately palliated with ipilimumab-nivolumab.