To the best of our knowledge, in this study, we present for the first time that the sequence of chemotherapy and toripalimab might influence the pCR in patients with locally advanced ESCC, with relatively manageable treatment-related AEs. The pCR was 36% in the experimental group and 7% in the control group, suggesting that the different sequence of administration of chemotherapy and toripalimab might affect the clinical efficacy. The regimen was well-tolerated, and the AEs of neoadjuvant chemoimmunotherapy were similar to those reported in the studies on advanced stage esophageal cancer[6, 14]. No new AEs occurred during the treatment. The most common AEs in the stage of neoadjuvant therapy were mainly caused by chemotherapeutic agents. One patient developed tropoin elevation during pulmonary infection after surgery, and this patient died from severe pulmonary infection. One patient withdrew from the study because of immune enteritis in the stage of neoadjuvant therapy. These findings suggested that we should pay much attention to the emergence of immune-related AEs, especially those uncommon AEs during treatment.
Studies conducted in patients with ESCC receiving neoadjuvant concurrent chemoradiotherapy showed that 18 of the 37 patients achieved pCR, and the rate of pCR was 49%, while the occurrence of AEs ≥ grade 3 was as high as 25%. Patients whose tumors achieved pCR to neoadjuvant therapy had significantly longer relapse-free survival and overall survival. Since the factors influencing therapeutic regimens for esophageal cancer between Eastern and Western countries are different, therapeutic strategies adopted in Asian countries, especically in Japan and China, often differ from those in Western countries. The JCOG9907 study compared the clinical efficacy of neoadjuvant chemotherapy and adjuvant chemotherapy in patients with stage II/III ESCC, the results showed that the 5-year survival rate of patients who received neoadjuvant chemotherapy was significantly higher than that of those who received adjuvant chemotherapy.There was no increase of postoperative complications and hospital mortality[4, 5]. Therefore, at present in Japan, patients with stage II/III ESCC are recommended for neoadjuvant chemotherapy followed by surgery or simultaneous chemoradiotherapy. In China, there was no standard regimen for patients with stage II/III ESCC, the recommended regimen for this kind of patients is to enroll in clinical studies.
In China a randomized clinical trial in patients with locally advanced ESCC comparing the neoadjuvant chemoradiotherapy to neoadjuvant chemotherapy showed that the pCRs were 35.7% and 3.8% individually, and there was no significant difference in postoperative morbidity between the two groups (47.4% vs 42.6%). The pCR rate of neoadjuvant chemotherapy in the Chinese population with locally advanced ESCC is relatively low and needs improvement. If the performance status of patients is good, simultaneous chemoradiotherapy might be a good choice. With the development of immunotherapy, clinical trials based on immunotherapy are also conducted in a neoadjuvant setting in such patients. In PALACE-1 study, pembrolizumab was added in chemoradiotherapy for ESCC, the rate of pCR was 55.6% (10/18), grade III and higher AEs occurred in 13 patients (13/20, 65%), and one patient died. Simultaneously, the addition of atezolizumab in neoadjuvant chemoradiotherapy in patients with resectable esophageal adenocarcinoma showed that the rate of pCR was 25%, grade 3/4 AEs was observed in 40% (16/40) patients and one patient died. The incidence of grade III and higher AEs in patients with locally advanced ESCC treated with neoadjuvant concurrent chemoradiotherapy combined with anti-PD-1/PD-L1 antibodies was high. The neoadjuvant therapeutic regimen containing immunotherapy in patients with locally advanced ESCC needs to be optimized. In Chinese patients, neoadjuvant chemoimmunotherapy might be one of the choices[4, 5, 17, 18].
When the anti-PD-1 antibody is combined with other kinds of therapies such as chemotherapy, radiotherapy, anti-angiogenesis, or immunomodulator, the sequence of anti-PD-1 antibody and other regimens might influence the clinical efficacy[7, 20–22]. The TONIC trial was a non-comparative phase II study to explore whether induction treatment could induce a more inflamed tumor microenvironment. Sixty-seven patients with metastatic triple-negative breast cancer were randomized to nivolumab without induction or with 2-week low-dose induction with irradiation or chemotherapy, the results showed that the ORRs in the doxorubicin cohort, cisplatin cohort, cyclophosphamide corhort, radiotherapy cohort, and no induction cohort were 35%, 23%, 8%, 8%, and 17%, respectively. Immune-related genes involved in PD-1/PD-L1 and T cell cytotoxicity pathways were upregulated in the doxorubicin and cisplatin cohorts. These results indicated that preconditioning with some chemotherapeutic agents could induce a more inflamed tumor microenviroment which would favor of immunotherapy. GeparNuevo study was a phase II study to investigate whether the addition of durvalumab into neoadjuvant chemotherapy increases the pCR rate in patients with early triple-negative breast cancer, the patients were divided into three cohorts, the pCR rate with durvalumab was 53.4% versus placebo 44.2% (P = 0.287), while in the window-phase cohort the pCR rate was 61% versus 41.4% (P = 0.048). This study showed again that preconditioning with durvalumab 2 weeks before chemoimmunotherapy could improve the pCR rate,which showed the importance of administration sequence of immunotherapy and chemotherapy. Therefore, when designing a chemoimmunotherapy regimen, the doses, sequence, and administration intervals should be considered. Considering the mechanisms of synergy between chemotherapy and immunotherapy, sequential approaches might be helpful.
The mechanisms of postponing the application of anti-PD-1 antibody for 2 days when combined with chemotherapy with anti-PD-1 antibody may be as follows: studies in preclinical animal models found that cyclophosphamide could upregulate PD-1 and PD-L1 expression on tumor-specific CD8 + T cells in tumor-bearing C3 mice.Then the application of anti-PD-1 treatment could increase the anti-tumor effect, thus increasing the systemic anti-tumor effect. Short-term chemotherapy in patients with triple negative breast cancer may induce a more favorable tumor microenvironment and increase the likelihood of response to PD-1 blockade therapy. In lung squamous cell cancer and colon cancer, there is a synergistic effect induced by low dose chemotherapy followed by anti-PD-1 antibodies[24, 25]. Therefore, theoretically, the effect of using PD-1 antibodies after chemotherapy may be better than that of using them simultaneously, but this does not imply that the later use of an anti-PD-1 antibody might yield better efficacy. Studies in patients with brain metastasis of melanoma have found that the immunotherapeutic efficacy within 4 weeks after stereotactic radiosurgery treatment is better than that after 4 weeks. The subgroup analysis of the PERCIFIC study also showed that the efficacy of anti-PD-L1 was better in the group in which anti-PD-L1 was applied within 2 weeks after the end of simultaneous chemoradiotherapy. A clinical study showed that the earlier application of anti-PD-1 antibody the better. Therefore, we postpone the application of anti-PD-1 antibody for 2 days after chemotherapy. When the anti-PD-1 antibody was applied alone, the T cells with anti-tumor activity in peripheral blood reached the peak 1 or 2 weeks after the first dose of anti-PD-1 antibody, and then decreased gradually. To reduce the influence of the second cycle of chemotherapy on activated T cells, the immune-chemotherapy cycle in the experimental group was 3–4 weeks. In this study, a trend of higher pCR rate was observed in the experimental cohort than that in control cohort (χ2 = 3.092, P = 0.079), indicating the importance of the sequence of chemotherapy and immunotherapy.
Thus far, there were no biomarkers for predicting the efficacy of anti-PD-1 therapy in esophageal cancer, although PD-L1 expression has been used to guide anti-PD-1 therapy in non-small cell lung cancer without driver gene mutations. In KEYNOTE 590 study, patients with PD-L1 CPS ༞10 could benefit more from chemotherapy combined with pembrolizumab, while the risk of death in patients with CPS < 10 could also be reduced. Patients could benefit from anti-PD-1 therapy regardless of the expression of PD-L1 in ATTRACTION-3 and ESCORT studies[14, 29]. In this study, we examined the expression of PD-L1 CPS in 14 patients (7 in each group) with the samples before treatment, the results showed that almost all the patients had PD-L1 CPS ≤ 1 (except one with PD-L1 CPS 10), which indicated that our patients might represent the group of patients insensitive to anti-PD-1 therapy. Two of the 6 patients with PD-L1 CPS ≤ 1 in the experimental cohort achieved pCR, and the rate of pCR in the experiment group was 36.4%, it might be even higher if we could avoid the influence of COVID-19. A higher rate of pCR might be obtained by delaying anti-PD-1 antibody application by 2 days when chemotherapy plus anti-PD-1 therapy were used.
There are several limitations of this study. First was the influence of COVID-19 from February to May in 2020. The chemoimmunotherapy cycles were forcibly prolonged, and the operations were postponed in several patients, which may be one of the reasons that the rate of pCR in this study was lower than that reported previously[18, 30–32]. Second, as this study was a small sample exploratory study, we initially planned to enroll 30 patients. Because the representation of the small sample was not strong, bias was inevitable in the process of selecting patients. The PD-L1 expression results showed that only one patient with PD-L1 CPS was 10, the others were PD-L1 ≤ 1, which suggested that the patients in this study may only present the group of patients insensitive to anti-PD-1 therapy. Thirdly, due to the limitation of the preoperative specimens, not all the patients have enough tissue specimens for PD-L1 CPS examination. Lastly, there was no exploratory study on other biomarkers that might affect the efficacy. However, our results showed that the pCR rate in patients with delayed application of anti-PD-1 antibody was 36.4%. There was a statistical boundary difference, suggesting that when chemotherapy combined with anti-PD-1 antibody was used in a neoadjuvant setting, delaying anti-PD-1 antibody therapy by 2 days might have greater clinical benefits.