Neoadjuvant toripalimab combined with gemcitabine and cisplatin in resectable locally advanced head and neck squamous cell carcinoma (NeoTGP01): An open label, single-arm, phase Ib clinical trial

Abstract

Background: Neoadjuvant programmed death receptor-1 (PD-1) inhibitors have drawn increasing attention in locally advanced head and neck squamous cell carcinoma (HNSCC). In this study, we investigated the safety and efficacy of gemcitabine and cisplatin (GP), combined with a PD-1 inhibitor, in patients with locally advanced HNSCC.

Materials and methods: A total of 23 eligible patients were administered two cycles of toripalimab and GP followed by surgical resection. The primary endpoints were safety, treatment-related adverse events (TRAEs), and non-operation delay rates. The secondary endpoints consisted of pathological complete response (pCR) rate, major pathological response (MPR) rate, objective response rate (ORR), and R0 resection rate.

Results: The incidence of TRAEs from grades 1 to 4 was 43.5%, 34.8%, 13.0%, and 8.7%, respectively. Grade 3/4 TRAEs included neutropenia, fatigue, hyperglycemia, nausea and vomiting, decreased appetite, rash, and diarrhea. No treatment-related surgical delay was observed. The radiographic response rates were 5.0% (CR), 40.0% (PR), and 55.0% (SD). The ORR reached 45.0%. Eighteen patients underwent successful surgical resection. The R0 resection rate was 100%. The pathological response rates were 16.7% (pCR), 27.8% (MPR, two of five near-pCR), 16.7% (PPR), and 38.8% (NPR). CD4, CD8, CD20, and CD38 expression in the tumors significantly increased after neoadjuvant chemotherapy. The increase in CD20 levels after neoadjuvant treatment in patients with pCR/MPR was significantly higher than in patients with PPR/NPR.

Conclusion: Triweekly neoadjuvant toripalimab-GP is feasible and achieves promising pCR and MPR rates in patients with resectable locally advanced HNSCC.

Trial registration: Chinese clinical trial registry, ChiCTR2100043743, Registered 27 Febrary 2021- Retrospectively registered, http://www.chictr.org.cn/showproj.aspx?proj=120570.

Keywords: Head and neck squamous cell carcinoma; Immunotherapy; Neoadjuvant treatment; Pathological response rate.

Fig. 1

Clinical trial flow diagram

Fig. 2

Waterfall plot of the characteristic of treatment response (n = 18). Each column indicates one patient, ranging from the highest to the lowest rate of pathological response. Corresponding sequence numbers of patients are labeled below. Two dashed horizontal lines denote 50% and 90% pathological responses. Three (16.7%) patients achieved pCR at the primary site, five (27.8%) achieved MPR (two among which were near-pCR, 1%), three (16.7%) achieved PPR, and seven (38.8%) were NPR

Fig. 3

Magnetic resonance images (T1 enhanced sequence images) and H&E staining images (20 ×) of primary site tumor before and after two courses of chemo-immunotherapy. Patient 6(A) and patient 7(B), suffering the squamous cell carcinoma at the mandible (T4aN1M0) and gingiva (T2N3bM0) respectively. As for patient 6, the maximum diameter of tumor was about 23 mm before neoadjuvant treatment, which was significantly enhanced on enhanced scan. After neoadjuvant treatment, the maximum diameter of tumor was still about 23 mm, and the intensity of enhancement was not significantly reduced. For patient 7, before neoadjuvant treatment, the maximum diameter of tumor was about 32 mm, which was also significantly enhanced on enhanced scan. Though the maximum diameter of tumor shrank to about 25 mm after neoadjuvant treatment, and the intensity of enhancement was significantly reduced, the radiological response was still assessed to be SD. Surprisingly, both of them achieved pCR at the primary site after two courses of chemo-immunotherapy. Histopathological assays of pre-treatment show numerous infiltrating tumor nest with nuclear enlargement, nuclear hyperchromasia, prominent nucleoli and mitotic figures, which conforms to the characteristics of cancer cells.After treatment, histopathology revealed interstitial fibrous tissue hyperplasia, infiltration of multinucleated giant cells and chronic inflammatory cells (including lymphocyte and plasma cells infiltration), and no residual tumor inside the tumor bed. Histopathology revealed interstitial fibrous tissue hyperplasia, infiltration of multinucleated giant cells (arrows) and chronic inflammatory cells (including lymphocyte and plasma cells infiltration), and no residual tumor inside the tumor bed

Fig. 4

Box plots of immunohistochemical evaluation. A–E The expression of CD4, CD8, CD20, CD38, and CD68 in tissues pre- and post- treatment. F The difference in CD20 expression pre- and post-treatment in patients with pCR/MPR and PPR/NPR. G Spearman’s correlation analysis of the difference in CD20 expression pre- and post-treatment with pathological response rates (p < 0.05, r = 0.75). (H–I) CD20 expression pre- and post-treatment in the patient with MPR (G) and NPR (I)

Fig. 5

Swimming plot of survival. Each lane indicates one patient, ranked by the survival time. Corresponding sequence numbers of patients are labeled on the left. Characteristics of patients are shown on the top. No disease progression and death happened until writing. The median follow-up from the first day of treatment was 10.5 months, ranging from 6 to 18 months