Concurrent Definitive Immunoradiotherapy for Patients with Stage III-IV Head and Neck Cancer and Cisplatin Contraindication

Jared Weiss, Siddharth Sheth, Allision M Deal, Juneko E Grilley Olson, Samip Patel, Trevor G Hackman, Jeffrey M Blumberg, Thomas J Galloway, Shetal Patel, Adam M Zanation, Colette J Shen, D Neil Hayes, Christopher Hilliard, Ranee Mehra, Karen P McKinnon, Hsing-Hui Wang, Mark Christian Weissler, Jessica R Bauman, Bhishamjit S Chera, Benjamin G Vincent, Jared Weiss, Siddharth Sheth, Allision M Deal, Juneko E Grilley Olson, Samip Patel, Trevor G Hackman, Jeffrey M Blumberg, Thomas J Galloway, Shetal Patel, Adam M Zanation, Colette J Shen, D Neil Hayes, Christopher Hilliard, Ranee Mehra, Karen P McKinnon, Hsing-Hui Wang, Mark Christian Weissler, Jessica R Bauman, Bhishamjit S Chera, Benjamin G Vincent

Abstract

Purpose: Although cisplatin plus radiotherapy is a standard treatment of locally advanced head and neck squamous cell carcinoma (LA-HNSCC), cisplatin contraindication is common. Radiation elicits and promotes tumor-directed immune stimulation, which may potentiate anti-PD-1 therapy. We provide the first efficacy report of combined pembrolizumab and definitive radiotherapy in LA-HNSCC.

Patients and methods: This single-arm, multi-institution, phase II study (NCT02609503) enrolled 29 cisplatin-ineligible patients. Patients received radiotherapy concurrently with three cycles of pembrolizumab 200 mg every 3 weeks followed by three adjuvant cycles. The primary endpoint was a progression-free survival (PFS) of ≥16 months. Correlative studies included peripheral blood flow cytometry and Luminex cytokine profiling.

Results: Reasons for cisplatin ineligibility included otopathy (69.0%), nephropathy (20.7%), and neuropathy (6.9%). With median follow-up of 21 months, estimated 24-month PFS and overall survival rates were 71% (95% confidence interval, 49%-84%) and 75% (51%-88%). The primary PFS endpoint has exceeded the hypothesis and its median has not been reached. Toxicities were typical of radiotherapy; however, high rates of grade 3/4 lymphopenia (58.6%) were observed. Flow cytometry revealed a relative decline in CD4 T cells and B cells, but not CD8 T cells. Upon treatment, frequencies of transitional B cells and tissue-like memory B cells increased, while resting memory B cells decreased. Patients with progression had greater percentages of baseline naïve B cells and fewer marginal zone B cells.

Conclusions: Pembrolizumab and radiotherapy is efficacious in LA-HNSCC and should be evaluated in a randomized trial. The observed changes in B-cell markers deserve further study both as potential biomarkers and as therapeutic targets.

©2020 American Association for Cancer Research.

Figures

Figure 1
Figure 1
A)PFS and B)OS in the total study population
Figure 2:
Figure 2:
Lymphopenia A) Absolute lymphocyte count over time B) Proportions of lymphocytes in periphery. Declined frequencies of memory/effector CD4 T cells at week 40 (p=0.0022); naïve CD4 T cells at week 20 and 40 respectively (p=0.0079 and p=0.0061). B cells were also decreased at week 20 (p=0.0033) though significantly recovered by week 40 (p=0.04) (2-way ANOVA and Tukey’s multiple comparisons tests were performed at α=0.05; error bar=SD).
Figure 3:
Figure 3:
Peripheral Blood Flow Cytometry A. Comparison of baseline lymphocyte phenotypes between progressors and non-progressors. Patients with progression had greater percentages of baseline naïve B cells (p=0.0077) and fewer marginal zone B cells (p=0.0088) (2-way ANOVA and Sidak’s multiple comparisons were determined by 6 of progressors and 13 of non-progressors; α=0.05; error bar= SD). Patients without progression are represented with open circles and patients with progression in closed circles. B. B cells subsets post XRT. Significantly increased transitional B cells at week 4 (p=0.0246), week 20 (p=0.0051), and week 40 (p=0.0008); increased tissue-like memory (TLM) B cells at week 20 (p=0.0173) and week 40 (p=0.0398). On the other hand, frequencies of resting memory (RM) B cells significantly declined at week 20 (p=0.0002) and week 40 (p=0.0097) (2-way ANOVA and Tukey’s multiple comparisons tests were performed at α=0.05; error bar=SD). C. Memory T cells subsets post XRT. PD1 expression was significantly down regulated at week 4 and 20 on both CD4 (p=0.0373 and p=0.0495) and CD8 T cells (p<0.0001). PD1 expression was significantly resumed on CD8 T cells at week 40 when patients were off treatment (p=0.0134). Frequencies of T regulatory cells showed no changes by looking into Foxp3 expression and CD25+CD127− phenotypes. Among all three exhaustion markers (CTLA4, TIGIT, and CD39), CD39 expression on CD4 memory T cells were up regulated at week 20 (p=0.0359) (2-way ANOVA and Tukey’s multiple comparisons tests were performed at α=0.05; error bar=SD). D. Naïve T cells subsets post XRT. Pembrolizumab showed limited effect on naïve T cells. Decreased PD1 expression was observed on naïve CD8 T cells at week 4 and 20 (p=0.0009 and 0.0178). The percentage of CD25 expression on CD4 naïve T cells was significantly increased at week 20 and week 40 comparing to baseline (p<0.0001 and p=0.0047). Once the patients were off treatment (week 40), the CD25 expression significantly dropped comparing to week 20 (p=0.0031) (2-way ANOVA and Tukey’s multiple comparisons tests were performed at α=0.05; error bar=SD).
Figure 4:
Figure 4:
Quality of Life A: Changes in Global and Functional indexes over time: Global, PWB, SFWB, EWB, FWB, HN B: Changes in individual HN subscales over time C: PROs

Source: PubMed

Подписаться