GENEVA – Level of programmed death-ligand 1 (PD-L1) expression does not impact patient-reported outcomes (PROs) among those receiving durvalumab for stage III non–small cell lung cancer (NSCLC), according to a retrospective analysis of the phase III PACIFIC trial.
The findings support durvalumab for all comers regardless of PD-L1 expression, reported lead author Marina Garassino, MD, of Fondazione IRCCS – Instituto Nazionale dei Tumori in Milan, who presented findings at the European Lung Cancer Congress.
The PACIFIC trial involved 713 patients with stage III NSCLC who did not progress after platinum-based concurrent chemoradiotherapy, demonstrating both improved progression-free and overall survival. Patients were randomized 2:1 to receive either durvalumab (10 mg/kg) or placebo IV every 2 weeks for up to 1 year. Out of 713 patients involved in the trial, 63% had PD-L1 tumor expression level data available for the present analysis, allowing for subgrouping into five categories: expression level of at least 25%, less than 25%, at least 1%, less than 1%, or unknown. The investigators compared PROs from these cohorts using the European Organisation for Research and Treatment of Cancer core quality of life questionnaire and lung cancer module (EORTC QLQ-C30 and -LC13). With scores ranging from 1 to 100 points, clinically meaningful differences were defined by score changes exceeding 10 points. Changes during treatment, from baseline to week 48, were analyzed by a mixed model for repeated measures, overall responses for improvement rates by logistic regression, and hazard ratios for time to deterioration (TTD) by a stratified Cox proportional-hazards model.
The investigators found that most PROs remained consistent over time, without clinically meaningful variations between PD-L1 expression levels. However, as with the entire PACIFIC treatment population, patients in the present analysis showed changes in some PROs. At the meeting presented by the European Society for Medical Oncology, Dr. Garassino noted that it would be unrealistic to describe all PRO comparisons; instead, she presented several examples. For one, in patients receiving durvalumab, dysphagia and alopecia improved in four out of five PD-L1 subgroups and all subgroups, respectively, while patients in the placebo arm reported improvements in both measures regardless of PD-L1 expression. Other improvements tended to favor the durvalumab group; for instance, compared with other subgroups, patients with PD-L1 expression of at least 25% were more likely to report improved chest pain, physical functioning, pain, emotional functioning, and hemoptysis. In contrast, patients receiving placebo with PD-L1 expression less than 25% were more likely to report improved cough. Still, the investigators concluded that the overall picture did not suggest major differences in PROs by PD-L1 expression level, noting that global quality of life did not differ, and symptom improvement rates and time-to-deterioration measures generally aligned with the intent-to-treat population, judging by overlapping 95% confidence intervals and hazard ratios.
“These data support the PACIFIC regimen for the standard of care for stage III unresectable non–small cell lung cancer patients,” Dr. Garassino concluded.
Invited discussant Fabrice Barlesi, MD, PhD, of Aix-Marseille University, said that studies such as this one are important to ensure that investigator-implemented measures of response are calibrated to patient experiences. As an example, Dr. Barlesi noted that many clinicians would say that grade 2 diarrhea is a completely manageable adverse event, but not all patients would agree.
In this light, Dr. Barlesi said that the present findings are valuable, but they are not without flaws. He noted that 11 out of 13 symptoms from the quality of life lung cancer module were not reported, and that one-third of patients in the PACIFIC trial lacked PD-L1 expression level data.
Considering these shortcomings, and more broadly, difficulties comparing patient-reported outcome studies because of various measurement techniques, Dr. Barlesi called for standardization.
“We need to standardize the analysis of quality of life data,” Dr. Barlesi said. “We should correct for the multiplicity of tests. … we should identify some specific quality of life outcomes that we want to look at in the protocol.” He continued to suggest a variety of ideal characteristics for studies evaluating patient-reported outcomes, including defined statistical measures and protocols for missing data.
Without a standardized approach, “cross trial comparison will be a nightmare for all of us,” Dr. Barlesi said.
The study was funded by AstraZeneca. The investigators reported financial relationships with Roche, BMS, Lilly, and others.
SOURCE: Garassino et al. ELCC 2019. .