Abstract
Background/Aim: The prognosis of recurrent/metastatic squamous cell carcinoma of the head and neck (R/M SCCHN) is poor, although immune checkpoint inhibitors (ICIs), such as nivolumab, have been shown to prolong survival. We investigated the factors that predict the efficacy of nivolumab when selecting an appropriate treatment strategy for patients with R/M SCCHN. Patients and Methods: Forty-four Japanese patients with R/M SCCHN treated with nivolumab between May 2017 and October 2021 were analyzed. The primary endpoint of the study was overall survival (OS). We defined pre-treatment tumor size (PTS) as the sum of the size of all measurable lesions, and tumor growth rate (TGR) as the total growth rate of the largest tumor diameter on CT scans taken to determine treatment response, divided by the interval between CT scans. Receiver operating characteristic (ROC) curves were used to identify the cutoff points of PTS and TGR for OS. Cox proportional hazards regression analysis was performed to examine the relationships between various factors, including patient characteristics, PTS, and TGR, as well as treatment outcomes. Results: In multivariate analysis, Eastern Cooperative Oncology Group (ECOG) performance status (PS) score ≥1, progressive disease (PD) as best overall response (BOR), and TGR >0.60%/day were independent risk factors for poor OS in patients with R/M-SCCHN. Conclusion: Higher TGR, poor PS, and PD as BOR may be prognostic factors in patients with R/M SCCHN.
- Recurrent/metastatic squamous cell carcinoma of the head and neck
- programmed cell death protein 1
- nivolumab
- pre-treatment tumor size
- tumor growth rate
Head and neck cancer is a malignant tumor that arises in the head and neck region, with squamous cell carcinoma (SCC) being the most common histological type. Most patients with squamous cell carcinoma of the head and neck (SCCHN) have advanced disease at the time of initial diagnosis, and more than half of the patients with advanced SCCHN develop local recurrence or distant metastasis within 3 years of initial treatment (1, 2). The prognosis of patients with unresectable recurrent/metastatic SCCHN (R/M SCCHN) is generally poor, especially for platinum-refractory R/M SCCHN, defined as recurrence or progression within 6 months of platinum-based chemotherapy, with no drug reported to improve survival in previous randomized controlled trials (3, 4). However, this situation regarding the treatment of patients with R/M SCCHN has changed since the development of immune checkpoint inhibitors (ICIs). The CheckMate 141 trial demonstrated that treatment with nivolumab, an anti-programmed cell death-1 (PD-1) agent, prolonged overall survival (OS) in patients with platinum-refractory R/M SCCHN compared with standard single-agent therapy (5). In addition, the KEYNOTE-048 trial showed that pembrolizumab, another anti-PD-1 antibody, as a monotherapy or in combination with chemotherapy (platinum plus 5-fluorouracil), improved OS compared to conventional treatment, particularly in R/M SCCHN patients with a positive combined positive score (CPS) for programmed cell death ligand 1 (PD-L1) (6). Based on these results, nivolumab is used in Japan as a first-line therapy for patients with platinum-refractory R/M SCCHN or as a second-line therapy for patients with platinum-sensitive R/M SCCHN. Pembrolizumab (combined with chemotherapy, such as 5-fluorouracil plus cisplatin or carboplatin) is used as the standard first-line therapy for patients with platinum-sensitive R/M SCCHN or without prior chemotherapy.
However, ICIs alone have lower response rates and longer time to response than regimens combining cytotoxic chemotherapy and molecularly targeted therapy, as demonstrated in the Checkmate141 trial that showed a response rate of only 13.3% to nivolumab in patients with platinum-refractory R/M SCCHN (5). Furthermore, attention should be paid to ICI-specific adverse events, known as immune-related adverse events (irAEs), when treatment with ICI is initiated.
Thus, identifying the factors that predict the efficacy of ICIs may be useful in selecting an appropriate treatment strategy for patients with R/M SCCHN; however, these factors have not been thoroughly discussed. Recently, several studies have reported that pre-treatment tumor size (PTS) or tumor growth rate (TGR) is a useful prognostic factor in the ICIs treatment of various malignancies (7-10). In this study, we investigated the significance of PTS and TGR as prognostic predictors in patients with R/M SCCHN treated with nivolumab in real-world clinical practice.
Patients and Methods
Study design. This single-center retrospective cohort study was conducted between May 2017 and October 2021 at the Yokohama City University Hospital in Yokohama, Japan. The medical records of patients with R/M SCCHN treated with nivolumab were reviewed retrospectively. The primary endpoint of the study was the OS of the patients. The requirement for informed consent was waived due to data anonymity (7, 11). The protocol was reviewed and approved by our institutional review board (approval ID: B201000038).
Patients. Patients were eligible if they were under 80 years old; had clinically confirmed recurrent and/or metastatic SCC of the nasopharynx, oral cavity, oropharynx, hypopharynx, larynx, sinonasal cavity, or cancers of unknown origin; had inoperable recurrent or metastatic lesions; received the nivolumab regimen; were examined for treatment response by computed tomography (CT) scans at least once; and had an Eastern Cooperative Oncology Group (ECOG) performance status (PS) of 0-3.
Medical records were retrospectively reviewed for the following characteristics: age, sex, primary tumor location, status of human papillomavirus infection with tumor p16 expression in patients with oropharyngeal SCC, ECOG PS prior to the use of the nivolumab regimen, multiple sites of lesions, lines of prior systemic cancer therapy not including cisplatin, platinum-refractory or not, radiotherapy history, date of diagnosis and treatment, last follow-up visit or death, histology.
The treatment response was evaluated in accordance with the Response Evaluation Criteria in Solid Tumors (RECIST version 1.1) using CT scans approximately every two months during the drug treatment period. According to RECIST 1.1, tumor responses were classified as the best overall response (BOR): complete response (CR), partial response (PR), stable disease (SD), or progressive disease (PD). Patients with lesions not measurable by CT scan with RECIST version 1.1 were excluded. All the patients included in this study were Japanese.
Treatment regimen. Patients were treated with either intravenous nivolumab (240 mg) every 2 weeks or nivolumab (480 mg) every 4 weeks. Nivolumab was administered until disease progression, intolerable toxicity, or a patient decision was made.
Assessment of tumor parameters. In this study, measurable tumor lesions were defined as lesions with the longest diameter >10 mm on CT scan, and measurable lymph nodes were defined as those with the shortest diameter >15 mm, as per RECIST version 1.1. However, patients with unmeasurable lesions, such as pleural effusion, ascites, bone lesions, or cancerous lymphangiopathy, were excluded from this study. Seventeen of the 61 patients with R/M SCCHN were excluded. A total of 44 patients with R/M SCCHN treated with nivolumab were enrolled in this study. PTS was defined as the sum of the sizes of all measurable lesions immediately before the start of nivolumab in individual patients. TGR was defined as the growth rate obtained by dividing the sum of the sizes of all measurable lesions on two CT scans immediately before nivolumab, divided by the interval between CT scans (12). We calculated the TGR (before nivolumab administration) assessed during the wash-out period before the introduction of nivolumab and the TGR after nivolumab assessed during the first cycle of treatment (i.e., between the introduction of nivolumab and the first evaluation).
Statistical analysis. A ROC curve was used to identify the cutoff point as the point nearest to the upper-left corner of the chart with the highest sensitivity and specificity for PTS and TGR for OS. OS was defined as the time interval from the first dose of nivolumab until death from any cause or the last confirmed survival. Fisher’s exact test was used to examine the correlations between categorical variables. Cox proportional hazard regression analysis was used to perform multivariate comparisons for categorical variables with p<0.05, using Fisher’s exact test. OS was analyzed using the Kaplan-Meier method with the Wilcoxon log-rank test. Patients who were lost to follow-up were censored at the date of last contact/follow-up. The Pearson’s correlation coefficient was used to assess the correlation between PTS and TGR before and after nivolumab administration. Statistical analysis was performed using JMP software (JMP Pro 15.0.0, SAS Institute Inc., Cary, NC, USA) and GraphPad Prism version 6.05 (GraphPad Software, San Diego, CA, USA). For all comparisons, statistical significance was set to p<0.05.
Results
Patient characteristics. The clinical characteristics of the 44 patients with R/M SCCHN are summarized in Table I. The median age at treatment initiation was 64.5 years (range=40-80 years). Most patients were men (88.6%). Thirty-six patients had an ECOG PS score of 0 (81.8%), seven patients had an ECOG PS score of 1 (15.9%), and one patient had an ECOG PS score of 3 (2.3%). None of the patients in this study had an ECOG PS score of 2. The primary sites of disease were as follows: hypopharynx (38.6%); oropharynx (22.7%), including six p16-positive cases, three p16-negative cases, and one unknown case; oral cavity (18.2%); sinonasal (9.1%); nasopharynx (6.8%), larynx (2.3%); and cancers of unknown primary (2.3%). Twenty-eight patients had multiple sites of lesions (63.6%). Nineteen patients had no line of prior systemic cancer therapy, excluding cisplatin (43.2%), 23 patients had one line of prior systemic cancer therapy (52.3%), and two patients had two lines of prior systemic cancer therapy (4.5%). Twenty-seven patients had platinum-refractory (61.4%). Thirty-six patients had radiotherapy history (81.8%). BOR was 6.8% CR (three patients), 15.9% PR (seven patients), 27.3% SD (12 patients), and 50.0% PD (22 patients).
Patient characteristics.
Cutoff values for PTS and TGR. The median PTS was 33.7 mm (range=11.5-148.5 mm), and the median TGR was 0.98%/day (range=−1.0-4.5 mm). The cutoff point of the PTS was 33.3 mm, with an area under the curve (AUC) of 0.828 [95% confidence interval (CI)=0.703-0.952] (Figure 1A). The PTS in 21 patients was ≤33.3 mm (47.7%). Similarly, the cutoff point of the TGR was 0.60%/day with an AUC of 0.811 (95%CI=0.673-0.95) (Figure 1B). The TGR of 15 patients was ≤0.60%/day (34.1%).
Receiver operating characteristic (ROC) curves for overall survival (OS) according to pre-treatment tumor size (A) and tumor growth rate (B). (A) The area under the curve (AUC) was 0.828 for pre-treatment tumor size, with a cutoff value 33.3. (B) The AUC was 0.811 for tumor growth rate with a cutoff value 0.60.
Treatment outcomes. The 1-year OS rate in the present study was 54% (95%CI=41-70) (Figure 2A), and the median survival time was 13.8 months, which was comparable to the results of a previous Checkmate141 study. Univariate analysis was performed to evaluate the prognostic significance of clinical and treatment characteristics for OS (Table II). We found that patients with ECOG PS of 1 and 3; PD as BOR; >33.3 mm in PTS; and >0.60%/day in TGR had a significantly shorter OS (p=0.037, p=0.0011, p=0.0036, and p=0.0001, respectively). In contrast, other parameters such as age, sex, multiple sites of lesions, number of lines of prior systemic cancer therapy, platinum-refractory status, and radiotherapy history were not significantly correlated with OS. The Fisher’s exact test showed that the number of patients with TGR >0.60%/day included significantly more patients with PD as BOR (p=0.010). Cox multivariate analysis was performed to identify independent prognostic factors for survival in patients with R/M SCCHN (Table II). We found that patients with ECOG PS of 1 and 3, PD as the BOR, and >0.60%/day in TGR had a significantly shorter OS (p=0.0003, p=0.035, and p=0.0094, respectively).
Kaplan-Meier curves for overall survival (OS) in the overall population (A), according to Eastern Cooperative Oncology Group Performance status (ECOG PS) (B), the best overall response (BOR) (C), pre-treatment tumor size (D), and tumor growth rate (E). CR: Complete response; PR: partial response; SD: stable disease; PD: progressive disease.
Univariate and multivariate analysis of overall survival.
Furthermore, Kaplan-Meier analysis revealed that OS was significantly longer in patients with ECOG PS of 0 than in those with ECOG PS of 1 and 3 (hazard ratio for death=0.12; 95%CI=0.020-0.71; p=0.0001, Figure 2B). OS was also significantly longer in patients with CR or PR, SD than in those with PD as the BOR (hazard ratio for death=0.21; 95%CI=0.098-0.46; p<0.0001, Figure 2C). In addition, OS was significantly longer in patients with ≤33.3 mm in PTS than in those with over 33.3 mm in PTS (hazard ratio for death=0.31; 95%CI=0.14-0.63; p=0.0017, Figure 2D). Lastly, OS was significantly longer in patients with ≤0.60%/day in TGR than in those with >0.60%/day in TGR (hazard ratio for death=0.15; 95%CI=0.092-0.40; p<0.0001, Figure 2E). Interestingly, there was almost no correlation between PTS and TGR (r=0.12, p=0.44; Figure 3A), and a weak positive correlation was found between TGR before and after nivolumab treatment (r=0.36, p=0.024; Figure 3B). These results suggest that the ECOG PS, BOR, and TGR may be independently associated with the survival of patients with R/M SCCHN treated with nivolumab.
Correlation between pre-treatment tumor size and tumor growth rate (A, r=0.12) and between tumor growth rate before and after nivolumab (B, r=0.36).
Discussion
We here demonstrated that TGR >0.60%/day was independently associated with worse survival in patients with R/M SCCHN treated with nivolumab in a single-center retrospective cohort study. We also confirmed that ECOG PS and BOR were useful in predicting the survival of these patients. Huang et al. demonstrated pharmacodynamic changes in circulating exhausted-phenotype CD8 T cells by immune profiling of peripheral blood from patients with melanoma before and after treatment with pembrolizumab. They reported that the relationship between CD8 T cells reinvigoration and tumor burden suggested a calibration of immune responses to the antigen burden and raised the possibility that even robust reinvigoration by anti-PD-1 therapy may be clinically ineffective if the tumor burden is high (13). As demonstrated in the present study, TGR was associated with tumor response to nivolumab treatment, which is consistent with a previous report (14). We observed that the TGR before and after nivolumab treatment showed a weak positive correlation, suggesting that patients with a high TGR immediately before nivolumab administration may have a lower response to nivolumab treatment. A larger TGR as an independent poor prognostic factor for patients with R/M SCCHN who were treated with nivolumab in this study is also consistent with previous results. Suzuki et al. reported the predictive significance of tumor burden and growth rate at baseline for the survival of patients with R/M SCCHN (8). These results are consistent with the consensus that ICIs generally respond poorly to rapidly growing tumors (15). Whereas we need to pay attention to atypical patterns of response such as pseudoprogression when treating patients with ICIs; several reports indicate that the rate of immunotherapy pseudoprogression is less than 2% in head and neck cancers (16). This may be because the immune surveillance mechanism of SCCHN is suppressed due to the decreased function of tumor-infiltrating lymphocytes, increased function of regulatory T cells, and over-expression of cancer antigens (17). Larger tumors could lead to dramatic changes in the immune system and harbor more immunosuppressive cells and molecules that dampen antitumor activity, suggesting that the tumor burden could be negatively correlated with the response to ICIs administration (18). Thus, ICIs administration may be an unfavorable option for treating patients with R/M SCCHN and rapid-growth tumors.
Similarly, PTS or baseline tumor size has been reported to be a prognostic factor in patients with several malignancies, including R/M SCCHN, treated with ICIs (7, 9, 10). However, in this study, the PTS was not an independent factor in predicting the survival of patients with R/M SCCHN treated with nivolumab. This discrepancy may be due to the retrospective design of this single-center cohort study, which included a small number of patients and a short follow-up period. On the other hand, PTS was not correlated with TGR in the present study, suggesting that TGR may better reflect disease status, such as dynamic changes in cancer cell proliferation of recurrent/metastatic lesions, compared to PTS.
In the present study, poor ECOG PS was also a poor prognostic factor for patients with R/M SCCHN who received ICI, which is consistent with the results of previous literatures (19-21). ICIs are considered less effective in patients with poor PS, and their status could be more immunosuppressive (21, 22). Patients with a low response rate to ICIs could benefit from prolonged survival with subsequent chemotherapy. However, patients with poor PS are generally not candidates for cytotoxic chemotherapy as salvage chemotherapy, resulting in no benefit from subsequent chemotherapy after ICI administration (23). This study also showed that R/M SCCHN patients with PD as a BOR had poorer survival, consistent with the results of previous studies demonstrating that a higher BOR was associated with improved survival in patients with R/M SCCHN or other malignancies (21, 24-26). Compared with patients with PD after nivolumab administration, patients with a favorable response to treatment are likely to fully benefit from these treatment effects, as they can receive subsequent therapy in addition to nivolumab.
PD-L1 expression is an established biomarker for the treatment of non-small cell lung cancer with ICIs (27). Specifically, in R/M SCCHN, the KEYNOTE-048 study demonstrated that positive PD-L1 expression enhances the therapeutic effect of pembrolizumab and that PD-L1 CPS, rather than the PD-L1 tumor proportion score (TPS), was useful for stratification to measure the therapeutic effect of pembrolizumab (6). In contrast, a longer follow-up in the exploratory subgroup analysis of the CheckMate-141 trial showed that nivolumab was beneficial regardless of PD-L1 expression (28). In this study, we did not evaluate the impact of PD-L1 CPS on survival, as CPS in patients with R/M SCCHN treated with nivolumab is not routinely examined in Japanese clinical practice. C-reactive protein (CRP) (29), neutrophil-to-lymphocyte ratio (NLR) (30), genetic factors such as soluble PD-L1 (sPD-L1) (31), circulating tumor DNA (ctDNA) in blood measured before and during ICI treatment (32), and the development of irAEs (33) may be useful prognostic markers for R/M SCCHN patients treated with ICI.
This study had several limitations. As mentioned previously, the present study had a retrospective design as a single-center cohort study with a small number of patients and a short follow-up period. In addition, the study included heterogeneous patient backgrounds, such as the primary site of disease, ECOG PS status, approach to prior systemic cancer therapy, platinum-refractory status, and radiotherapy history. The cut-off values for the PTS and TGR may be affected by many factors, such as patient characteristics and the number of patients included in the study. Therefore, our findings need to be validated in future multi-institutional studies with longer follow-up periods and larger patient populations.
Conclusion
In summary, we performed a single-center retrospective cohort study in a real-world clinical setting to investigate candidate prognostic factors for treatment outcomes in patients with R/M SCCHN who were treated with nivolumab. R/M SCCHN patients with a good PS, higher BOR, and lower TGR may respond to nivolumab and achieve longer survival. These results may aid in the selection of treatment options for patients with R/M SCCHN.
Acknowledgements
The Authors thank Motohiko Tokuhisa, MD, who was involved with data collection.
Footnotes
Authors’ Contributions
Conceptualization, D.S.; Methodology, D.S.; Validation, D.S. and N.O.; Formal Analysis, Y.H., K.K., and D.S.; Investigation, D.S., Y.H., K.K., and M.I.; Resources, D.S., M.I., H.T., and Y.I.; Data Curation, D.S., Y.H., K.K., and M.I.; Writing – Original Draft Preparation, Y.H.; Writing – Review and Editing, D.S. and N.O.; Visualization, Y.H. and D.S.; Supervision, N.O.
Conflicts of Interest
The Authors declare that there are no conflicts of interest in relation to this study.
- Received July 11, 2023.
- Revision received August 10, 2023.
- Accepted August 12, 2023.
- Copyright © 2023, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.