Lymphocyte-Sparing with Hypofractionated Proton Radiation in Locally Advanced Non-Small Cell Lung Cancer (LA-NSCLC)

Abstract

Adjuvant immunotherapy is now the standard of care for LA-NSCLC, but severe lymphopenia has been shown to blunt its efficacy. Conventionally fractionated photon radiation (cfRT) has been associated with high rates of post-treatment lymphopenia and elevated neutrophil-to-lymphocyte ratio (NLR) in LA-NSCLC, which have been linked to worse survival. We hypothesized hypofractionated proton radiation (hpRT) may reduce the risk of post-treatment lymphopenia and elevated NLR. This was a secondary analysis of an institutional prospective phase I trial conducted from 2015-2016 in which twenty patients with LA-NSCLC were treated with hpRT (52.5–60.0 CGE over 15 daily fractions). Primary endpoints have been reported previously. Patients were compared to a historical cohort (2001-2017) of 212 LA-NSCLC patients treated with photon cfRT (52–75 Gy over 24-37 fractions). All patients received concurrent platinum and taxane-based chemotherapy. Patients were included if they had a complete blood count at pre-RT, during-RT, and 4 months post-RT start. An inverse probability of treatment weighting (iPTW) approach was used to match patients on age, sex, histology, clinical stage, and use of consolidation chemotherapy. Rates of grade ≥3 lymphopenia were compared using iPTW-weighted odds ratios. Risk of local failure, distant failure, and death were evaluated by weighted Cox regression. Patients in the hpRT group were stage IIB (5%), IIIA (75%), and IIIB (20%). Patients in the cfRT group were stage IIB (1.5%), IIIA (68.5%), and IIIB (30%). Consolidation chemotherapy after definitive chemoRT was given to 16(80%) and 99(47%) of hpRT and cfRT patients, respectively. The median follow up for patients alive in the hpRT group was 34 months (range 16-46) and for the cfRT group was 37 months (range 1-132). The rates of grade ≥3 lymphopenia at pre-RT, during-RT, and post-RT timepoints for hpRT and cfRT groups, respectively, were 0% vs. 3% (p=0.41), 65% vs. 92% (p<0.001), and 5% vs. 24% (p=0.06). Compared to hpRT, the IPTW-adjusted odds ratio (OR) for cfRT developing grade ≥3 lymphopenia during RT was 6.11 (95% CI 2.03-17.9); for post-RT, OR=8.01 (95% CI 1.03-62.6). The NLR was significantly higher in the cfRT group compared to the hpRT group during-RT (adjusted estimate 9.26, 95% CI 2.20-16.31) and post-RT (adjusted estimate 5.45, 95% CI 1.14-9.75). Compared to hpRT, the IPTW-adjusted hazard ratio of local failure, distant failure, and death for cfRT was 2.13 (95% CI 0.76-5.98), 2.89 (95% CI 0.96-8.69), and 1.59 (95% CI 0.91-2.80), respectively. Compared to cfRT, hpRT was associated with less grade ≥3 lymphopenia and lower NLR during and after RT with similar tumor control and survival compared to historical controls. This data supports future trials of lymphocyte-sparing hpRT in combination with immunotherapy for the treatment of LA-NSCLC.