Induction chemotherapy backbone in frail patients with advanced NSCLC treated with chemotherapy plus pembrolizumab: a single institution retrospective audit of dose intensities from modified regimens.

Guidelines historically recommended mono-chemotherapy for the 1st line treatment of elderly patients with non-small cell lung cancer (NSCLC) and poor performance status (PS). Nowadays, there is no clear indication whether chemo-immunotherapy (chemo-IO) combinations can be effectively delivered in this population. We collected induction chemotherapy data in consecutive patients with advanced NSCLC treated with carboplatin-based chemotherapy regimens plus pembrolizumab, to compute the received dose intensity (RDI) from standard regimens or patient-tailored regimens modified due to age, comorbidities and PS. Comorbidities were stratified according to the comorbidity-polypharmacy score (CPS). The established cut-off of ≥85% for RDI was used to define adequate delivery. 116 pts were treated from Feb-20 to July-23, of whom 96 and 20 with non-squamous and squamous NSCLC, treated with carboplatin-pemetrexed or carboplatin-paclitaxel doublets plus pembrolizumab, respectively. The majority of patients were aged ≥ 70 years (52.6%), the median CPS was 5, with 58.6% having a CPS ≥5, whilst 47.4%, 44.8% and 7.8% had an Eastern Cooperative Oncology Group (ECOG) - PS of 0, 1 and 2, respectively. PD-L1 TPS were <1% in 31.9% and 1-49% in 65.4%. Overall, 47.4% received a priori modified regimens due to poor PS, age, or comorbidities. Among patients with non-squamous NSCLC, median received doses of carboplatin and pemetrexed were 1.37 AUC/week and 138.8 mg/m2/week, with RDIs of 86% and 75% (p < 0.01) for patients treated with standard or modified regimens, respectively. Of note, the RDI was 57.9% among patients with ECOG-PS 2. However, patients treated with modified regimens experienced similar toxicities as those treated with standard regimens, despite being older (p < 0.01), with higher PS (p < 0.01) and more comorbid (p = 0.03). Patients treated with modified regimens achieved a shorter survival (7.1 vs 13.9 months), which is comparable to IO-free historical controls. Among patients with squamous NSCLC, 90% received modified regimens upfront, with median received doses of carboplatin and paclitaxel of 1.19 AUC/week and 40 mg/m2/week, and an overall RDI of 73.5%. Although regimen modifications ensure a safe administration of chemotherapy plus pembrolizumab in frail patients, the RDI seems to be subtherapeutic, especially in those with squamous histology. Dedicated trials are needed to implement combination strategies in this population.