Research ArticleClinical Studies

Real-world Experience With Pembrolizumab for Advanced-stage Head and Neck Cancer Patients: A Retrospective, Multicenter Study

TAKAFUMI NAKANO, RYUJI YASUMATSU, KAZUKI HASHIMOTO, RYOSUKE KUGA, TAKAHIRO HONGO, HIDETAKA YAMAMOTO, MIOKO MATSUO, TAKAHIRO WAKASAKI, RINA JIROMARU, TOMOMI MANAKO, SATOSHI TOH, MUNEYUKI MASUDA, MORIYASU YAMAUCHI, YUICHIRO KURATOMI, MASAHIKO TAURA, TORANOSHIN TAKEUCHI and TAKASHI NAKAGAWA
Anticancer Research July 2022, 42 (7) 3653-3664; DOI: https://doi.org/10.21873/anticanres.15854

Abstract

Background/Aim: This study investigated the effectiveness of pembrolizumab with or without chemotherapy on advanced-stage head and neck cancer (HNC), including nasopharyngeal, sinonasal cavity and external auditory canal cancer, in a real-world setting. Patients and Methods: We retrospectively collected data from 97 HNC patients who were treated with pembrolizumab alone (n=60) or with chemotherapy (n=37), and we investigated the association between clinicopathological findings and treatment response or prognosis. Results: Patients treated with pembrolizumab and chemotherapy had a 1-year overall survival (OS) of 72.8%, objective response rate (ORR) of 48.6%, and serious (≥G3) adverse events (AEs) of 29.7%. Patients treated with pembrolizumab alone had a 1-year OS of 51.9%, ORR of 21.7%, and ≥G3 AEs of 6.7%. Both the ORR and disease control rate (DCR) in the pembrolizumab with chemotherapy group were significantly better than those in the pembrolizumab group (p=0.074 and p=0.00101, respectively). Among patients with distant metastasis, patients on pembrolizumab with chemotherapy achieved significantly better OS than pembrolizumab alone (p=0.0039). Among patients in the pembrolizumab group, both AE-positive and better performance status were associated with longer OS (p=0.011 and p=0.0037, respectively). Conclusion: Our real-world experience reinforces the durability and effectiveness of pembrolizumab for HNC patients. Additionally, our results suggest that pembrolizumab with chemotherapy might be recommended for patients with distant metastasis and no prior treatment. Further studies are needed to determine the optimal treatment strategy for HNC.

Key Words:

Advanced-stage head and neck cancer (HNC) still has a poor prognosis. Lately, immune checkpoint inhibitors (ICIs), including pembrolizumab have been shown to provide a significant clinical survival benefit for patients with advanced-stage head and neck squamous cell carcinoma (1). In keeping with these findings, the US Food and Drug Administration (FDA) has approved pembrolizumab for use in the treatment of HNC. In the phase 3 KEYNOTE-048 study, pembrolizumab alone or in combination with chemotherapy (carboplatin or cisplatin and 5-fluorouracil) improved overall survival (OS) compared to cetuximab with chemotherapy. However, this study focused on the effect on hypopharyngeal, laryngeal, oral cavity, and oropharyngeal cancer, and did not include other sites of HNC such as the nasopharynx, sinonasal cavity, or eternal auditory canal. In addition, there has been no large series in a real-world clinical setting that has shown an association between prognosis and adverse events (AEs) in HNC patients treated with pembrolizumab and chemotherapy. In the present study we, therefore, retrospectively investigated the effectiveness of pembrolizumab with or without chemotherapy on advanced-stage HNC, including nasopharyngeal, sinonasal cavity and external auditory canal cancer.

Patients and Methods

Case selection. We retrospectively collected data of 97 HNC patients treated with pembrolizumab with or without chemotherapy from the Kyushu University Hospital, Kyushu National Cancer Center, Fukuoka University Hospital, Saga University Hospital, Kyushu Medical Center, Kitakyushu Municipal Medical Center, Hamanomachi Hospital, and Sasebo Kyosai Hospital. All cases were diagnosed as squamous cell carcinoma from the head and neck region. We classified the efficacy of treatment as complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD) (2). The objective response rate (ORR) is defined as the proportion of patients with a CR and PR, and the disease control rate (DCR) as the proportion of patients with an ORR and SD. We also assessed toxicity according to the Common Terminology Criteria for Adverse Events (CTCAE) version 4.0. This study was approved by the Institutional Review Board at Kyushu University (approval number: 2021-222).

Programmed death ligand-1 (PD-L1) expression. PD-L1 immunohistochemical staining was performed using biopsied or surgically resected formalin-fixed paraffin-embedded tissues and the primary antibodies (22C3; PharmDx). The combined positive score (CPS) was calculated as the number of PD-L1-stained cells (tumor cells, lymphocytes, and macrophages) divided by the total number of viable tumor cells multiplied by 100.

Statistical analyses. The JMP Statistical Discovery software (ver. 16.0; SAS, Cary, NC) was used for all analyses. We used Fisher’s exact test to evaluate the significance of associations between variables. OS and progression free survival (PFS) were calculated with the Kaplan-Meier method, and the differences were compared using the log-rank test. A p-value <0.05 was considered statistically significant.

Results

Clinicopathological findings. Characteristics of patients in this study are summarized in Table I. All patients in this study ranged in age from 30 to 90 years (median=68 years). Most patients were male (n=84, 86.6%). Eastern Cooperative Oncology Group (ECOG) performance status (PS) was 0-1 in 83 (85.6%) and 2 in 14 (14.4%) patients. The primary tumor site was the oral cavity in 19 (19.6%), nasopharynx in 7 (7.2%), oropharynx in 20 (20.6%), hypopharynx in 33 (34.0%), larynx in 9 (9.3%), sinonasal cavity in 4 (4.1%), and other sites in 5 (5.2%) patients; the other sites consisted of 4 patients with external auditory canal and 1 patient with salivary gland cancer. The disease status was distant metastasis in 24 (24.7%) and locoregional recurrence in 73 (75.3%) patients. Pembrolizumab was first-line treatment in 88 (90.7%), second-line treatment in 6 (6.2%), and third-line treatment in 3 (3.1%) patients. The CPS was <1 in 8 (8.2%), 1-19 in 46 (47.4%), and >20 in 43 (44.3%) patients. Sixty-six (68.0%) patients discontinued pembrolizumab due to either progressive disease (PD, n=50, 51.5%), adverse events (AEs, n=13, 13.4%), or other reasons (n=3, 3.1%). The other 31 (32.0%) patients continued pembrolizumab.

View this table:
Table I.

Characteristics of patients.

Sixty (61.9%) of 97 patients were treated with pembrolizumab alone. The patients treated with pembrolizumab alone ranged in age from 39 to 90 years (median=68 years), and were predominantly male (n=53, 88.3%). ECOG PS was 0-1 in 50 (83.3%) and 2 in 10 (16.7%) of these patients. The primary site was the oral cavity in 14 (23.3%), nasopharynx in 4 (6.7%), oropharynx in 11 (18.3%), hypopharynx in 19 (31.7%), larynx in 6 (10.0%), sinonasal cavity in 2 (3.3%), and other sites in 4 (6.7%) patients. The disease status at the initial visit to our hospital among the 60 patients who had not been previously treated was distant metastasis in 6 (10.0%), and locoregional and/or distant recurrence in 54 (90.0%) patients. Pembrolizumab was first-line treatment in 55 patients (91.7%), the second-line treatment in 4 patients (6.7%), and third-line treatment in 1 patient (1.7%). The CPS was <1% in 3 (5.0%), 1-19% in 28 (46.7%), and >20% in 29 (48.3%) of the 60 patients. Forty-three (71.7%) patients discontinued pembrolizumab due to either PD (n=33, 55.0%) or AE (n=10, 16.7%). The other 17 (28.3%) patients continued pembrolizumab.

The remaining 37 (39.1%) of 97 patients were treated with pembrolizumab in combination with chemotherapy containing cisplatin/carboplatin and 5-fluorouracil. The patients treated with this combination therapy ranged in age from 30 to 82 years (median, 68 years), and were also predominantly male (n=31, 83.8%). ECOG PS was 0-1 in 33 (89.2%) and 2 in 4 (10.8%) of these 37 patients. The primary site was the oral cavity in 5 (13.5%), nasopharynx in 3 (8.1%), oropharynx in 9 (24.3%), hypopharynx in 14 (37.8%), larynx in 3 (8.1%), and sinonasal cavity in 2 (5.4%) patients, and another site in 1 patient (2.7%). The disease status at the initial visit to our hospital among the 37 patients who had not been previously treated was distant metastasis in 18 (48.6%), and locoregional and/or distant recurrence in 19 (51.4%). Pembrolizumab was the first-line therapy in 33 (89.2%), the second-line therapy in 2 (5.4%), and the third-line therapy in 2 (5.4%) of the 37 patients. The CPS was <1% in 5 (13.5%), 1-19% in 18 (48.6%), and >20% in 14 (37.8%) patients.

Figure 1 illustrates the number of patients by CPS for each primary site. The only patients with a CPS <1 were those with oropharynx (n=2), hypopharynx (n=5), or larynx (n=1) cancer. Patients with a CPS 1-19 or CPS>20 were seen in all primary sites.

Figure 1.
Figure 1.

Bar graphs illustrate the number of patients by combined positive score (CPS) for each primary site. Light orange bars represent the CPS <1 subgroup. Orange bars represent the CPS 1-19 subgroup. Dark orange bars represent the CPS ≥20 subgroup.

Twenty-three (62.2%) of the 37 patients treated with pembrolizumab plus chemotherapy discontinued this combination therapy due to either PD (n=17, 45.9%), AE (n=3, 8.1%), or other reasons (n=3, 8.1%). The other 14 (37.8%) patients continued this combination therapy.

Table II shows the choice of platinum compound in this combination therapy. Twenty-nine (78.4%) patients were treated with 1 to 6 (average 3.24) cycles of cisplatin, and 8 (21.6%) were treated with 2 to 5 (average 3.1) cycles of carboplatin.

View this table:
Table II.

Choice of cisplatin or carboplatin for patients treated with pembrolizumab with chemotherapy (n=37).

Efficacy of pembrolizumab with/without chemotherapy. Best overall response (BOR) for each treatment group is shown in Table III, and for each CPS status is shown in Table IV.

Among the overall patient group, a CR was seen in 9 (9.3%), a PR in 22 (22.7%), an SD in 29 (29.9%), and a PD in 37 (38.1%) patients. The ORR was 32.0% and the DCR was 61.9%.

View this table:
Table III.

Best overall response (n=97).

View this table:
Table IV.

Best overall response according to CPS (n=97).

As shown in Table III, among patients treated with pembrolizumab alone (n=60), CR was seen in 6 (10.0%), PR in 7 (11.7%), SD in 18 (30.0%), PD in 29 (48.3%), ORR in 13 (21.7%), and DCR in 31 (51.7%) patients. In addition, among patients treated with pembrolizumab and chemotherapy (n=37), CR was seen in 3 (8.1%), PR in 15 (40.5%), SD in 11 (29.7%), PD in 8 (21.6%), ORR in 18 (48.6%), and DCR in 29 (78.4%) patients. Patients treated with pembrolizumab and chemotherapy showed a significantly better ORR (p=0.0074) and DCR (p=0.00101) than those treated with pembrolizumab alone.

As shown in Table IV, among patients with a CPS of >20 tumors (n=43), the BOR in the 29 patients treated with pembrolizumab alone was as follows: CR in 3 (10.3%) patients, PR in 4 (13.8%), SD in 7 (24.1%), or PD in 15 (51.7%) patients; the ORR was 24.1%, and DCR was 48.3%. In addition, BOR in 14 patients treated with pembrolizumab in combination with chemotherapy was as follows: CR in 1 (7.1%), PR in 4 (28.6%), SD in 7 (50.0%), and PD in 2 (14.3%) patients; the ORR was 35.7% and DCR was 85.7%. Also, patients treated with pembrolizumab plus chemotherapy showed a significantly higher DCR than those treated with pembrolizumab alone (p=0.0233).

Among patients with 1<CPS<19 tumors (n=46), the BOR in the 28 patients treated with pembrolizumab alone were as follows: CR in 3 (10.7%) patients, PR in 3 (10.7%), SD in 11 (39.3%), and PD in 11 (39.3%) patients; the ORR was 21.4%, and DCR was 60.7%. In addition, the BOR in the 18 patients treated with pembrolizumab in combination with chemotherapy was as follows: CR in 1 (5.6%), PR in 10 (55.6%), SD in 3 (16.7%), and PD in 4 (22.2%) patients; the ORR was 61.1% and DCR was 77.8%. Also, patients treated with pembrolizumab plus chemotherapy showed a significantly higher ORR than those treated with pembrolizumab alone (p=0.0271).

Among patients with a CPS <1 tumor (n=8), the BOR in all 3 patients treated with pembrolizumab alone was PD. In addition, the BOR in the 5 patients treated with pembrolizumab in combination with chemotherapy was as follows: CR in 1 (20.0%), PR in 1 (20.0%), SD in 1 (20.0%), and PD in 2 (40.0%) patients; the ORR was 40.0% and DCR was 60.0%. There were no statistically significant differences in ORR or DCR between patients treated with pembrolizumab alone and those treated with pembrolizumab plus chemotherapy (p=0.4642 and p=0.1964, respectively).

The CPS status did not affect the ORR or DCR (data not shown) in either of the treatment groups.

Adverse events and infusion reactions of pembrolizumab with/without chemotherapy. The profiles of AEs and infusion reactions are summarized in Table V. For patients overall, any grade AEs were reported in 28.3% and 40.5% (p=0.2678) of patients treated with pembrolizumab alone and pembrolizumab with chemotherapy, respectively, grade 3 or higher AEs were reported for 6.7% and 29.7% (p=0.0035) of patients in these two treatment groups. The AEs of all grades were pneumonitis (n=6, 10.0%), hypothyroidism (n=5, 8.3%), hypophysitis (n=1, 1.7%), hepatitis (n=1, 1.7%), nephritis (n=1, 1.7%), fatigue (n=1, 1.7%), hyponatremia (n=1, 1.7%), and infusion reaction (n=1, 1.7%) in patients treated with pembrolizumab alone, and pneumonitis (n=4, 10.8%), neutropenia (n=4, 10.8%), nephritis (n=2, 5.4%), hypophysitis (n=1, 2.7%), hepatitis (n=1, 2.7%), fatigue (n=1, 2.7%), febrile neutropenia (n=1, 2.7%), and other (n=1, 2.7%) in patients treated with pembrolizumab with chemotherapy. With regard to grade 3 or higher AEs, hypophysitis (n=1, 2.7%), pneumonitis (n=1, 2.7%), hyponatremia (n=1, 2.7%), and infusion reaction (n=1, 2.7%) were reported in patients treated with pembrolizumab alone, and neutropenia (n=4, 10.8%), pneumonitis (n=3, 8.1%), hypophysitis (n=1, 2.7%), fatigue (n=1, 2.7%), febrile neutropenia (n=1, 2.7%), and other (n=1, 2.7%) in patients treated with pembrolizumab with chemotherapy.

View this table:
Table V.

Profiles of adverse events and infusion reactions.

Prognostic analyses with pembrolizumab with/without chemotherapy, PS, CPS, and AE. As shown in Figure 2A and B, the overall 1-year OS rate was 58.8%, and the overall 1-year PFS rate was 20.2%. The 1-year OS rate was 51.9% in the patients treated with pembrolizumab alone and 72.8% in those treated with pembrolizumab with chemotherapy (Figure 2C). Patients treated with pembrolizumab with chemotherapy tended to have more favorable OS compared to those treated with pembrolizumab alone, but the difference was not statistically significant (Figure 2C, p=0.1336). The 1-year PFS rate was 20.1% in the patients treated with pembrolizumab alone and 16.4% in those treated with pembrolizumab with chemotherapy (Figure 2D, p=0.2331). Among patients with distant metastasis who had previously been untreated, patients treated with pembrolizumab and chemotherapy had a significantly better OS (Figure 2E, p=0.0039) and tended to have a better PFS (Figure 2F, p=0.075) than those treated with pembrolizumab alone.

Figure 2.
Figure 2.

Kaplan-Meier analysis for overall survival (OS) (A) and progression-free survival (PFS) (B) in all 97 head and neck carcinoma patients. Kaplan-Meier analysis for OS (C and E) and PFS (D and F) in head and neck carcinoma patients treated with pembrolizumab with or without chemotherapy. Patients treated with pembrolizumab with chemotherapy (1-year 72.8%) showed a trend toward better OS than those with pembrolizumab alone (1-year 51.9%), but the difference was not statistically significant (C) (p=0.1336). There was no difference in PFS between these two groups (D) (p=0.2331). Among patients with distant metastasis, those treated with pembrolizumab with chemotherapy had a significantly better OS (E) (p=0.0039) and a slightly but not significantly better PFS (F) (p=0.075) than those with pembrolizumab alone.

Among patients treated with pembrolizumab alone, patients with good PS (ECOG PS 0-1) had significantly better OS and PFS compared to those with poor PS (ECOG PS 2) (Figure 3A and B, p=0.0037 and p=0.0004, respectively). The 1-year OS rate was 56.8% in patients with good PS, but it was 25.0% in those with poor PS (Figure 3A). Among patients treated with pembrolizumab combined with chemotherapy, patients with good PS tended to have better OS compared to those with poor PS, but not significantly so (Figure 3C, p=0.09). In addition, there was no notable difference in PFS between patients with good and those with poor PS (Figure 3D, p=0.8146).

Figure 3.
Figure 3.

Kaplan-Meier analysis for overall survival (OS) and progression-free survival (PFS) in head and neck carcinoma patients treated with pembrolizumab alone (A and B) or pembrolizumab with chemotherapy (C and D) with respect to performance status (PS). Among patients treated with pembrolizumab alone, those with good PS (0-1) compared to poor PS (2) were significantly correlated with better OS (A) (p=0.0037) and PFS (B) (p=0.0004). Among patients treated with pembrolizumab and. chemotherapy, those with good PS showed a trend toward better OS than those with poor PS, but the difference was not statistically significant (C) (p=0.09). There was no difference in PFS between the two groups (D) (p=0.8146).

With respect to CPS, there was no significant difference in OS or PFS between patients with and without chemotherapy in the treatment regimen (Figure 4).

Figure 4.
Figure 4.

Kaplan-Meier analysis for overall survival (OS) and progression-free survival (PFS) in head and neck carcinoma patients treated with pembrolizumab alone (A and B) or with chemotherapy (C and D). There was no association between either OS (A and C) or PFS (B and D) band the combined positive score (CPS) status irrespective of treatment group. In all four panels, red lines represent the CPS <1 subgroup; green lines represent the CPS 1-19 subgroup; and blue lines represent the CPS ≤20 subgroup.

Figure 5A, B, C, and D shows the prognostic impacts of AEs. Among patients treated with pembrolizumab alone, those who developed AEs had a significantly longer OS (Figure 5A, p=0.0110), and a somewhat longer PFS (Figure 5B, p=0.2670). In contrast, among patients treated with pembrolizumab with chemotherapy, patients who developed AEs had a significantly shorter PFS (Figure 5D, p=0.0449) and a somewhat shorter OS (Figure 5C, p=0.2939).

Figure 5.
Figure 5.

Kaplan-Meier analysis for overall survival (OS) and progression-free survival (PFS) in head and neck carcinoma patients treated with pembrolizumab alone (A and B) or with chemotherapy (C and D). Among patients treated with pembrolizumab alone, the presence of adverse events (AE+) was significantly associated with a favorable OS (A) (p=0.0110) and tended to be correlated with a favorable PFS (B) (p=0.2670). Among patients treated with pembrolizumab with chemotherapy, the presence of AEs (AE+) was significantly associated with an unfavorable PFS (D) (p=0.0449) and tended to be correlated with an unfavorable OS (C) (p=0.2939). In all four panels, red lines represent the presence of AEs (AE+) subgroup, and blue lines represent the absence of AEs (AE-) subgroup. Kaplan-Meier analysis of OS (E and G) and PFS (F and H) in head and neck carcinoma patients treated with pembrolizumab alone (E and F) or with pembrolizumab plus chemotherapy (G and H) according to the best overall response (BOR). A complete response (CR) or partial response (PR) was significantly associated with OS (E) and PFS (F) among patients treated with pembrolizumab alone, and with PFS among patients treated with pembrolizumab with chemotherapy (H) and showed a nonsignificant trend toward an association with better OS among patients treated with pembrolizumab with chemotherapy (G).

Among patients overall, there was no specific trend in the BOR among primary sites (Figure 6A). Among patients treated with pembrolizumab alone, a CR was seen in patients with oral cavity, nasopharyngeal, oropharyngeal, and laryngeal cancer (Figure 6B). Among patients treated with pembrolizumab and chemotherapy, a CR was seen in patients with sinonasal cavity, oropharyngeal, and hypopharyngeal cancer (Figure 6C).

Figure 6.
Figure 6.

Bar graphs illustrate the percentages of all patients (A), patients treated with pembrolizumab alone (B), and patients treated pembrolizumab plus chemotherapy (C). The blue bar graphs represent the percentages of patients with a complete response (CR), the checked bar graphs are the percentages of patients with a partial response (PR), the dotted pattern bar graphs are the percentages of patients with stable disease (SD), and the brank pattern bar graphs are the percentages of patients with progressive disease (SD).

As for BOR, CR or RP was significantly associated with better OS (Figure 5E, p=0.0022) and PFS (Figure 5F, p=0.0006) than SD or PD among patients treated with pembrolizumab alone. All patients with a CR or PR were alive during the observation period (Figure 5E). In addition, a CR or PR in patients treated with pembrolizumab and chemotherapy was significantly correlated with a better DFS than SD or PD (Figure 5H, p=0.015), and patients treated with this regimen also showed a better OS, although the difference was not significant (Figure 5G, p=0.1034).

Discussion

In this retrospective study, our overall patient group, which consisted of patients treated with pembrolizumab alone and patients treated with pembrolizumab plus chemotherapy, showed a 1-year OS of 58.8% and 1-year PFS of 20.2% in a real-world setting. The patients treated with pembrolizumab alone showed a 1-year OS and 1-year PFS of 51.9% and 20.1%, respectively, while these values were 72.8% and 16.4% among patients treated with pembrolizumab and chemotherapy. Our study included patients with cancer of the nasopharyngeal, sinonasal cavity, and external auditory canal regions, and all these cancers were excluded from the KEYNOTE-048 clinical trial (1). In this clinical trial, the 1-year OS was 49% in patients treated with pembrolizumab alone and 55% in those treated with pembrolizumab and chemotherapy. Kim et al. (3) reported that the median OS was 11.8 months and the median PFS was 4.3 months in nasopharyngeal cancer patients treated with ICIs. Hashimoto et al. (4) reported that the 1-year OS and PFS in patients with salivary gland carcinoma treated with ICIs were 57.8% and 25.0%, respectively. Overall, these previous reports were in line with our results in the real-world. As in previous reports (1, 3, 4), the OS was better than the PFS in our results. This means that the effect of pembrolizumab lasted, or that pembrolizumab may increase the effect of next-line treatment.

The key finding of this study is that, among patients who had distant metastasis at the time of their first visit, those who were subsequently treated with pembrolizumab and chemotherapy showed a significantly better OS than those who were treated with pembrolizumab alone (Figure 2E). Since these patients had not been treated before, they had a good PS (PS 0-1) at the start of treatment, and this may have contributed to their favorable OS. In fact, patients with a good PS had a better OS than those with a poor PS (Figure 3C). On the other hand, the KEYNOTE-048 clinical study (1) investigated the effect of pembrolizumab with or without chemotherapy on patients with locoregional recurrence or distant metastasis. The patients with distant metastasis in the KYENOTE-048 clinical study included both patients with distant metastatic recurrence and those with distant metastasis at initial diagnosis. As a result, the patient group in the KEYNOTE-048 study was not exclusively limited to patients with distant metastasis without any prior treatment. Therefore, our study was able to contribute the recommendation that pembrolizumab with chemotherapy might be appropriate for patients with distant metastasis and good PS without any prior treatment.

Interestingly, we demonstrated that CPS status did not affect either prognosis (OS and PFS) or BOR (ORR and DCR) irrespective of treatment groups (Figure 4 and Table IV). In contrast, the KEYNOTE-048 clinical trial (1) showed that the median OS was 14.8 months in patients with a CPS >20, 10.8 months in patients with a CPS of 1-19, and 7.9 months in patients with a CPS <1 among patients treated with pembrolizumab alone, and that the median OS was 14.7 months in patients with a CPS >20, 12.7 months in patients with a CPS of 1-19, and 11.3 months in patients with a CPS <1 among patients treated with pembrolizumab and chemotherapy. Additionally, Ito et al (5) reported that there was a strong association between CPS and tumor proportion score (TPS) and that OS and PFS were better among patients with high TPS. In our study, all patients with a CPS <1 in the group treated with pembrolizumab alone showed a PD within 4 months (4 month-PFS of 0%), and a slightly but not significantly worse OS than the other groups (Figure 4A and B). In addition, there was no significant difference in OS and DFS between patients with a CPS <1 and those with a CPS >1 in this study (data not shown). Since the number of patients in our study was limited, further studies in a larger number of patients are needed.

As for the 4 cases of external auditory canal cancer, 2 of them had a CPS >20 and the others had a CPS <1. A previous study showed that PD-L1 protein expression was significantly associated with prognosis, and all patients treated with ICI died of disease (6). Similarly, all 4 patients with external auditory canal cancer in our study had a BOR of PD, and all 4 died of disease. In contrast, Noda et al. evaluated the effect of and ICI (nivolumab) on 9 patients with external auditory canal cancer and showed that the BOR was PR in 1 (11.1%) and SD in 2 (22.2%), and also showed that patients treated with nivolumab had a longer 1-year OS than those not treated with nivolumab (7). As the numbers of patients in the previous studies and our present analysis were small, further studies with a larger number of patients are needed.

In the comparison between treatment regimens, patients treated with pembrolizumab plus chemotherapy had a significantly better ORR or DCR than those treated with pembrolizumab alone (Table III). According to CPS, the KEYNOTE-048 clinical study showed that the ORR in patients treated with pembrolizumab and chemotherapy was 42.9%, 29.3%, or 30.8% in patients with a CPS >20, 1-19, or<1, respectively, and that the ORR in those treated with pembrolizumab alone was 23.3%, 14.5%, or 4.5% in patients with a CPS >20, 1-19, or <1 (1). Therefore, our real-world experiences were consistent with this clinical trial, except in the case of our patients with a CPS <1 who were treated with pembrolizumab alone. In addition, among patients with a CPS >20, patients with pembrolizumab and chemotherapy had a significantly better DCR than those with pembrolizumab alone (85.7% and 48.3%, respectively). Among patients with a CPS 1-19, pembrolizumab combination therapy showed a significantly better ORR than pembrolizumab monotherapy (61.1% and 21.4%, respectively). Overall, combination therapy achieved a better response than monotherapy. As for prognosis, patients with pembrolizumab and chemotherapy tended to have a better OS than those with pembrolizumab alone, but not to a statistically significant degree (Figure 2C).

Intriguingly, the impact of AEs on prognosis differed between the treatment regimens. To the best of our knowledge, our study is the first to evaluate the association between AEs and prognosis in patients with HNC treated with pembrolizumab and chemotherapy. We found that AEs in patients treated with pembrolizumab alone were a significantly better prognostic factor than non-AEs (Figure 5A). Similarly, previous studies in melanoma and lung cancer showed an association between AEs and better prognosis (8-10). As for incidence of AEs, previous studies showed that 58.3% of patients treated with pembrolizumab alone had AEs of any grade and 13.0-16.7% had AEs of grade 3 or higher (1, 11). In our study, 28.3% of patients had AEs of any grade and 6.7% had AEs of grade 3 or higher (Table V). In contrast, we also showed that patients with AEs in the pembrolizumab plus chemotherapy group trended to have a worse prognosis than those without non-AEs (Figure 5C and D). The incidence of AEs in the pembrolizumab with chemotherapy group was 40.5% for any grade and 29.7% for grade 3 or higher, which was higher than the incidence in the pembrolizumab alone group but less than that in the KEYNOTE-048 clinical study (95.3% and 71.0%. respectively). Chemotherapy (cisplatin/ carboplatin and 5-fluorouracil) in this combination treatment group might have an adverse effect. Therefore, the higher risk of AEs by this combination therapy in patients with recurrent or metastatic HNC patients has limited the treatment options for clinicians with patients having higher PS, who tend to develop AEs. Pembrolizumab alone might be a better treatment option for such patients, even though in our study the patients with higher PS still showed a worse prognosis in our study (Figure 3).

In conclusion, we retrospectively investigated the effectiveness of pembrolizumab alone or with chemotherapy on advanced HNC and demonstrated that patients treated with pembrolizumab and chemotherapy had a significantly better ORR or DCR depending on CPS than those treated with pembrolizumab alone. Additionally, among patients with distant metastasis at the initial diagnosis, patients treated with pembrolizumab and chemotherapy had a significantly better OS than those treated with pembrolizumab alone (p=0.0039). Therefore, our results demonstrated that pembrolizumab with chemotherapy is useful in patients with HNC, especially those with distant metastasis, no prior treatment and good PS. Further real-world experience or analysis in a larger number of patients and longer follow-up period are needed.

Acknowledgements

The English usage in this article was reviewed by KN international (http://www.kninter.com/).

Footnotes

  • ↵* These Authors contributed equally to this study.

  • Authors’ Contributions

    Conception and design: T. Nakano, R. Yasumatsu. Acquisition of data: T. Nakano, R. Yasumatsu, H. Yamamoto, R. Jiromaru, T. Manako, M. Masuda, M. Yamauchi, Y. Kuratomi, M. Taura, T. Takeuchi. Analysis and interpretation of data: T. Nakano, R. Yasumatsu, K. Hashimoto, M. Matsuo, T. Wakasaki, S. Toh, M. Masuda. Writing of the manuscript: T. Nakano, R. Yasumatsu. Administrative, technical, or material support: T. Nakano, R. Yasumatsu, K. Hashimoto, R. Kuga, T. Hongo, H. Yamamoto, M. Matsuo, T. Wakasaki, R. Jiromaru, T. Manako, S. Toh, M. Masuda, M. Yamauchi, Y. Kuratomi, M. Taura, T. Takeuchi. Study supervision: T. Nakano, R. Yasumatsu, T. Nakagawa.

  • Conflicts of Interest

    The Authors declare that there are no conflicts of interest to disclose.

  • Funding

    This study funded by JSPS (Japan Society for the Promotion of Science) KAKENHI (grant JP20K09757 to R.Y. and grant JP21K16845 to T.N.).

  • Received April 29, 2022.
  • Revision received May 20, 2022.
  • Accepted May 24, 2022.

References

  1. ↵
    2. Burtness B ,
    3. Harrington KJ ,
    4. Greil R ,
    5. Soulières D ,
    6. Tahara M ,
    7. de Castro G Jr ,
    8. Psyrri A ,
    9. Basté N ,
    10. Neupane P ,
    11. Bratland Å ,
    12. Fuereder T ,
    13. Hughes BGM ,
    14. Mesía R ,
    15. Ngamphaiboon N ,
    16. Rordorf T ,
    17. Wan Ishak WZ ,
    18. Hong RL ,
    19. González Mendoza R ,
    20. Roy A ,
    21. Zhang Y ,
    22. Gumuscu B ,
    23. Cheng JD ,
    24. Jin F ,
    25. Rischin D and KEYNOTE-048 Investigators
    : Pembrolizumab alone or with chemotherapy versus cetuximab with chemotherapy for recurrent or metastatic squamous cell carcinoma of the head and neck (KEYNOTE-048): a randomised, open-label, phase 3 study. Lancet 394(10212): 1915-1928 , 2019. PMID: 31679945. DOI: 10.1016/S0140-6736(19)32591-7
    OpenUrlCrossRefPubMed
  2. ↵
    2. Eisenhauer EA ,
    3. Therasse P ,
    4. Bogaerts J ,
    5. Schwartz LH ,
    6. Sargent D ,
    7. Ford R ,
    8. Dancey J ,
    9. Arbuck S ,
    10. Gwyther S ,
    11. Mooney M ,
    12. Rubinstein L ,
    13. Shankar L ,
    14. Dodd L ,
    15. Kaplan R ,
    16. Lacombe D and
    17. Verweij J
    : New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 45(2): 228-247, 2009. PMID: 19097774. DOI: 10.1016/j.ejca.2008.10.026
    OpenUrlCrossRefPubMed
  3. ↵
    2. Kim H ,
    3. Kwon M ,
    4. Kim B ,
    5. Jung HA ,
    6. Sun JM ,
    7. Lee SH ,
    8. Park K and
    9. Ahn MJ
    : Clinical outcomes of immune checkpoint inhibitors for patients with recurrent or metastatic head and neck cancer: real-world data in Korea. BMC Cancer 20(1): 727, 2020. PMID: 32758163. DOI: 10.1186/s12885-020-07214-4
    OpenUrlCrossRefPubMed
  4. ↵
    2. Hashimoto K ,
    3. Yasumatsu R ,
    4. Kuga R ,
    5. Hongo T ,
    6. Yamamoto H ,
    7. Matsuo M ,
    8. Wakasaki T ,
    9. Jiromaru R ,
    10. Manako T ,
    11. Toh S ,
    12. Masuda M ,
    13. Yamauchi M ,
    14. Kuratomi Y ,
    15. Uryu H ,
    16. Nakashima T ,
    17. Tamae A ,
    18. Tanaka R ,
    19. Taura M ,
    20. Takeuchi T ,
    21. Yoshida T and
    22. Nakagawa T
    : Treatment efficacy of PD-1 inhibitor therapy in patients with recurrent and/or metastatic salivary gland carcinoma. Anticancer Res 42(2): 981-989, 2022. PMID: 35093898. DOI: 10.21873/anticanres.15558
    OpenUrlAbstract/FREE Full Text
  5. ↵
    2. Ito T ,
    3. Okamoto I ,
    4. Tokashiki K ,
    5. Sato H ,
    6. Okada T ,
    7. Yamashita G ,
    8. Nagao T ,
    9. Hirai H ,
    10. Saigusa N and
    11. Tsukahara K
    : PD-L1 expression and survival rates using TPS and CPS for nivolumab-treated head-and-neck cancer. Anticancer Res 42(3): 1547-1554, 2022. PMID: 35220251. DOI: 10.21873/anticanres.15628
    OpenUrlAbstract/FREE Full Text
  6. ↵
    2. Hongo T ,
    3. Kuga R ,
    4. Miyazaki M ,
    5. Komune N ,
    6. Nakano T ,
    7. Yamamoto H ,
    8. Koike K ,
    9. Sato K ,
    10. Kogo R ,
    11. Nabeshima K ,
    12. Oda Y and
    13. Nakagawa T
    : Programmed death-ligand 1 expression and tumor-infiltrating lymphocytes in temporal bone squamous cell carcinoma. Laryngoscope 131(12): 2674-2683, 2021. PMID: 34143491. DOI: 10.1002/lary.29689
    OpenUrlCrossRefPubMed
  7. ↵
    2. Noda T ,
    3. Komune N ,
    4. Yasumatsu R ,
    5. Tsuchihashi NA ,
    6. Tamae A ,
    7. Matsumoto N ,
    8. Sato K ,
    9. Uchi R ,
    10. Koike K ,
    11. Wakasaki T ,
    12. Tanaka R and
    13. Nakagawa T
    : Therapeutic effect of Nivolumab for advanced/recurrent temporal bone squamous cell carcinoma. Auris Nasus Larynx 47(5): 864-869, 2020. PMID: 32446629. DOI: 10.1016/j.anl.2020.03.011
    OpenUrlCrossRefPubMed
  8. ↵
    2. Eggermont AMM ,
    3. Kicinski M ,
    4. Blank CU ,
    5. Mandala M ,
    6. Long GV ,
    7. Atkinson V ,
    8. Dalle S ,
    9. Haydon A ,
    10. Khattak A ,
    11. Carlino MS ,
    12. Sandhu S ,
    13. Larkin J ,
    14. Puig S ,
    15. Ascierto PA ,
    16. Rutkowski P ,
    17. Schadendorf D ,
    18. Koornstra R ,
    19. Hernandez-Aya L ,
    20. Di Giacomo AM ,
    21. van den Eertwegh AJM ,
    22. Grob JJ ,
    23. Gutzmer R ,
    24. Jamal R ,
    25. Lorigan PC ,
    26. Krepler C ,
    27. Ibrahim N ,
    28. Marreaud S ,
    29. van Akkooi A ,
    30. Robert C and
    31. Suciu S
    : Association between immune-related adverse events and recurrence-free survival among patients with stage III melanoma randomized to receive pembrolizumab or placebo: a secondary analysis of a randomized clinical trial. JAMA Oncol 6(4): 519-527, 2020. PMID: 31895407. DOI: 10.1001/jamaoncol.2019.5570
    OpenUrlCrossRefPubMed
    2. Haratani K ,
    3. Hayashi H ,
    4. Chiba Y ,
    5. Kudo K ,
    6. Yonesaka K ,
    7. Kato R ,
    8. Kaneda H ,
    9. Hasegawa Y ,
    10. Tanaka K ,
    11. Takeda M and
    12. Nakagawa K
    : Association of immune-related adverse events with nivolumab efficacy in non-small-cell lung cancer. JAMA Oncol 4(3): 374-378, 2018. PMID: 28975219. DOI: 10.1001/jamaoncol.2017.2925
    OpenUrlCrossRefPubMed
  10. ↵
    2. Morimoto K ,
    3. Yamada T ,
    4. Takumi C ,
    5. Ogura Y ,
    6. Takeda T ,
    7. Onoi K ,
    8. Chihara Y ,
    9. Taniguchi R ,
    10. Yamada T ,
    11. Hiranuma O ,
    12. Morimoto Y ,
    13. Iwasaku M ,
    14. Kaneko Y ,
    15. Uchino J and
    16. Takayama K
    : Immune-related adverse events are associated with clinical benefit in patients with non-small-cell lung cancer treated with immunotherapy plus chemotherapy: a retrospective study. Front Oncol 11: 630136, 2021. PMID: 33833990. DOI: 10.3389/fonc.2021.630136
    OpenUrlCrossRefPubMed
  11. ↵
    2. Cohen EEW ,
    3. Soulières D ,
    4. Le Tourneau C ,
    5. Dinis J ,
    6. Licitra L ,
    7. Ahn MJ ,
    8. Soria A ,
    9. Machiels JP ,
    10. Mach N ,
    11. Mehra R ,
    12. Burtness B ,
    13. Zhang P ,
    14. Cheng J ,
    15. Swaby RF ,
    16. Harrington KJ and KEYNOTE-040 investigators
    : Pembrolizumab versus methotrexate, docetaxel, or cetuximab for recurrent or metastatic head-and-neck squamous cell carcinoma (KEYNOTE-040): a randomised, open-label, phase 3 study. Lancet 393(10167): 156-167, 2019. PMID: 30509740. DOI: 10.1016/S0140-6736(18)31999-8
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Reprints and Permissions
Real-world Experience With Pembrolizumab for Advanced-stage Head and Neck Cancer Patients: A Retrospective, Multicenter Study
TAKAFUMI NAKANO, RYUJI YASUMATSU, KAZUKI HASHIMOTO, RYOSUKE KUGA, TAKAHIRO HONGO, HIDETAKA YAMAMOTO, MIOKO MATSUO, TAKAHIRO WAKASAKI, RINA JIROMARU, TOMOMI MANAKO, SATOSHI TOH, MUNEYUKI MASUDA, MORIYASU YAMAUCHI, YUICHIRO KURATOMI, MASAHIKO TAURA, TORANOSHIN TAKEUCHI, TAKASHI NAKAGAWA
Anticancer Research Jul 2022, 42 (7) 3653-3664; DOI: 10.21873/anticanres.15854
Twitter logo Facebook logo Mendeley logo

Jump to section

Keywords