The Role of Thoracic Vertebral Body Dosimetry in Minimizing Acute Hematologic Toxicities of Patients With Non-Small Cell Lung Cancer Receiving Lung Radiation Therapy and Immunotherapy

Abstract

IntroductionHematologic toxicities (HTs) are among the most common toxicities of combined immunotherapy and radiotherapy (RT). It remains essential to prevent RT-induced HTs because they can cause treatment discontinuation (influencing antitumoral effects), and because lymphopenia might dampen the effects of immunotherapy. To date, there are no studies examining the impact of thoracic vertebral body (TVB) RT dose on HTs in non-small cell lung cancer (NSCLC) patients receiving combined lung RT and PD(L)-1 immunotherapy. MethodsFor standardization, all doses were reported as 2-Gy equivalents (EQD2). Mirroring publications prior to the immunotherapy era, TVB volumes referred to T1-T10 and specific dosimetric parameters (DmeanEQD2, V5EQD2‒V60EQD2) were analyzed. Logistic regression estimated associations between grade ≥3 HTs (HT3+) and dosimetric/clinical parameters. Normal tissue complication probability (NTCP) models were constructed by logistic regression analysis modeling for HT3+. Receiver operating characteristic (ROC) analysis delineated TVB dosimetric thresholds, stratification of which was able to evaluate post-RT absolute lymphocyte count (ALC) and immunotherapy responses. Areas under the curve (AUCs) for NTCP models were corroborated by bootstrapping (optimism-corrected) methodology. ResultsIn 132 patients, there were 26 (19.7%) instances of HT3+. On multivariate analysis, DmeanEQD2 and V5EQD2‒V20EQD2 associated with HT3+ (P<0.05 for all). The NTCP models illustrated a 50% probability of HT3+ at a DmeanEQD2=39.8 Gy, V5EQD2=87.4%, V10EQD2=77.0%, and V20EQD2=68.4%. ROC analysis delineated optimal thresholds of HT3+ with DmeanEQD2 ±30.2 Gy, V5EQD2 ±69.1%, V10EQD2 ±64.6%, and V20EQD2 ±53.5%. Patients treated with values above those cutoffs had over double the risk of HT3+, with significant differences in post-RT ALC and immunotherapy responses (P<0.05 for all). AUCs for each individual parameter ranged from 0.743 to 0.798, and combining all four aforementioned cutoffs into a ROC curve resulted in a qualitatively higher AUC (0.836). ConclusionsThis hypothesis-generating work suggests that TVB dosimetry may equate with HT3+ in NSCLC patients undergoing combined lung RT/immunotherapy. Applying TVB dose constraints in this population could reduce HT3+ and avoid dampening of immunotherapy responses, but prospective validation is required.