3698 - Relationship between Elective Neck Irradiation and Severe Lymphopenia in Patients Treated with Head and Neck Chemoradiotherapy

T. A. Mickel¹, D. H. Moon², R. Hughes¹, B. Sumer¹, J. L. Shah³, V. Avkshtol¹, N. L. Pham Jr⁴, and D. J. Sher²; ¹University of Texas Southwestern Medical Center, Dallas, TX, ²Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, TX, ³Department of Radiation Oncology, University of Michigan, Ann Arbor, MI, ⁴Department of Radiation Oncology, UT Southwestern Medical Center, Dallas, TX

Purpose/Objective(s): Radiation-induced lymphopenia (RIL) has been associated with decreased survival in head and neck squamous cell carcinoma (HNSCC) and may reflect radiation (RT)-induced immunosuppression that can hinder response to immunotherapy. Previous institutional phase II trials have shown that de-escalating elective neck irradiation (ENI) is oncologically sound and results in superb quality-of-life. In this retrospective study, we investigate the relationship between different ENI dosing schemes and severe RIL in HNSCC patients receiving chemoradiotherapy (CRT). Materials/

Methods: This cohort included patients with newly diagnosed HNSCC of the oropharynx, larynx, and hypopharynx treated on the phase II INRT-AIR and INFIELD trials or with standard-of-care (SOC) CRT. In the INFIELD study, the ENI dose was 40 Gy in 20 fractions, treating only involved plus the adjacent nodal stations. Elective nodal irradiation (ENI) was excluded in INRT-AIR. The absolute lymphocyte count (ALC) was gathered from labs collected before treatment, at the time of lymphocyte nadir, and at 1-3 months, 6-9 months, and 1-2 years following initiation of treatment. Ipsilateral and contralateral carotid arteries (CA) and cervical vertebral bodies (CVB) were contoured and their dose metrics were recalculated and dichotomized at the median. Differences in ALC and severe lymphopenia (SL, grade 4) were compared between cohorts, and the relationships between SL and dose to the CA and CVB were interrogated.
Results: In total, 167 patients from INRT-AIR (N=61), INFIELD (N=59), and SOC (N=47) were included in the analysis. The risk of SL significantly increased with increasing ENI dose, as the probability of SL was 2%, 17% and 33% in INRT-AIR, INFIELD and SOC cohorts, respectively (p<0.001). The average Dmean (Gy) for the INRT-AIR, INFIELD and SOC cohorts was, respectively, 39.6, 50.7, and 56.8 for ipsilateral CA, 28.5, 33.3, and 51.4 for contralateral CA, and 32.7, 35.8, and 40 for CVB. The mean ALC nadir was significantly lower in patients with higher doses to the ipsilateral CA (343 vs. 437), contralateral CA (350 vs. 429, p=0.06), and CVB (337 vs. 443), p<0.05 for all other comparisons. The risks of SL were significantly greater with higher mean doses to the ipsilateral CA (25.6% vs. 6.1%), contralateral CA (29.6% vs. 2.4%), and CVB (29.3% vs. 2.4%), p<0.01 for all comparisons. On multivariable regression, both higher dose to the contralateral CA (odds ratio [OR] 7.0, p=0.02) and CVB (OR 6.6, p=0.02) were associated with SL. These results were replicated when using V40 as the dosimetry metric.
Conclusion: The extent of ENI in general, and the resulting dose to the contralateral CA and CVB, were significantly associated with SL during CRT for HNSCC. De-escalation of ENI may facilitate the success of immunotherapy regimens in patients receiving CRT for HNSCC.