Post-study immunotherapy confounds primary survival outcome

An exploratory analysis of the phase 3 MYSTIC trial suggests that the high rate of subsequent post-study immunotherapy received by patients with metastatic non-small cell lung cancer (NSCLC) who previously received chemotherapy, confounded the primary overall survival (OS) outcome.

Durvalumab (anti-PD-L1) is approved for the treatment of unresectable, stage 3 NSCLC. In phase 2 and 3 trials, durvalumab has shown clinical activity in heavily pre-treated patients with metastatic NSCLC.

Durvalumab vs chemotherapy

MYSTIC is an open-label, phase 3 study evaluating first-line durvalumab with or without tremelimumab vs platinum-based chemotherapy in metastatic NSCLC. A previous analysis showed a clinically meaningful improvement in OS with durvalumab vs chemotherapy in patients with high PD-L1 expression (PD-L1 in tumour cells of ≥25%; HR 0.76; P=0.036).

Subsequent immunotherapy

In the current analysis, Dr Niels Reinmuth (Asklepios Lung Clinic, Munich-Gauting, Germany) described subsequent treatment patterns and explored the effect of subsequent immunotherapy on the OS outcome with durvalumab vs chemotherapy. A markedly higher proportion of patients in the chemotherapy arm than in the durvalumab arm received subsequent immunotherapy. Among patients who received subsequent treatment, immunotherapy was administered to 14% of patients in the durvalumab group and 67% of patients in the chemotherapy group. This policy may have confounded the primary OS outcome.

Three models

To evaluate the effect of subsequent (post-study) immunotherapy on the OS data, three statistical models were employed in exploratory analyses:

  • the rank preserving structural failure time (RPSFT) method;
  • the inverse probability of censoring weighting (IPCW) method; and
  • a 2-stage method. The 2-stage model was the most appropriate method for evaluating the effect of imbalances in subsequent immunotherapy. Increased OS benefit with first-line durvalumab vs chemotherapy was observed after adjusting for the effect of subsequent immunotherapy (hazard ratio 0.66).
  1. Reinmuth N. Effect of post-study immunotherapy (IO) on overall survival (OS) outcome in patients with metastatic (m) NSCLC treated with first-line durvalumab (D) vs chemotherapy (CT) in the phase III MYSTIC study. ELCC 2019, Geneva, Switzerland; abstract LBA4.

The content and interpretation of these conference highlights are the views and comments of the speakers/authors.