Weekly paclitaxel, carboplatin and cetuximab (PCC) combination followed by nivolumab in platinum-sensitive recurrent and /or metastatic squamous cell carcinoma of head and neck: a retrospective analysis from two institutions in India [version 1; peer review: awaiting peer review]

Background: First line (1L) TP-Ex-like regimen followed by 2nd-line (2L) immunotherapy represents one of the standards of


Introduction
Squamous cell carcinomas of head and neck (SCCHN) accounts for 90% of all head and neck cancers (HNCs) 1,2 The cumulative risk of developing HNC (oral and pharynx) in India is 1 in 60, with males at higher risk than females (cumulative risk 1 in 41: 1 in 112). 3 Approximately 66% of HNCs in India are diagnosed at a locally advanced (LA) stage. 3 Though, a disease of sixth or seventh decade of life, HNC is increasingly being seen in younger population. 1 SCCHN is a difficult to treat cancer, especially in India, due to difference in genetic and etiological factors, in additional to logistic issues, compared to western world. 2 The etiology of HNC in India is primarily driven by smokeless tobacco chewing versus human papilloma virus (HPV) infection seen in western world. 2 In recurrent and/or metastatic (R/M) setting, treatment of SCCHN is even more difficult with poor outcomes.
Historically, in R/M SCCHN, combination chemotherapy regimens such as platinum plus 5-fluorouracil (5-FU) [4][5][6][7] and platinum plus taxane [8][9][10][11][12][13][14] significantly improved response rates compared with single agent chemotherapy, but they were not successful in demonstrating an improved OS and showed higher toxicity. 15 Cetuximab based combinations with platinum-based doublet chemotherapy such as the EXTREME regimen (cetuximab-cisplatin/carboplatin-5-fluorouracil   16,17 or the TPExtreme regimen (cetuximab-docetaxel-cisplatin), 18 followed by maintenance cetuximab until disease progression, showed increased OS compared with doublet chemotherapy and became the best 1L approach in R/M SCCHN. 19,20 Also the EXTREME regimen is associated with hematological toxicities that raised the need for safer regimens. 21 The CEMET trial showed that paclitaxel, carboplatin and cetuximab (PCC) regimen is a feasible 1L treatment option and is associated with significantly lesser toxicities than the EXTREME regimen (40% vs. 60%; p = 0.034). 22 Many patients with R/M HNSCC show disease progression after 1 st line platinum-based therapies. 23 In these patients, the U.S. Food and Drug Administration (FDA) has approved PD-1 inhibitors like nivolumab and pembrolizumab for 2 nd line treatment. 24,25 After the success of the Keynote-048 study, 26 1 st -line immuno-chemotherapy (IC) with pembrolizumab-platinum-5FU combination (irrespective of programmed death-ligand 1 [PD-L1] status) or immunotherapy alone with pembrolizumab (for PD-L1 ≥ 1%) became the preferred option of systemic therapy for platinum sensitive R/M SCCHN. 15 However, the EXTEME and TP-Extreme regimens, still remain one of the standard first-line options in R/M SCCHN, especially if immunotherapy is not feasible, in PD-L1 negative patient population, in oral primary, and in rapidly progressive bulky disease which requires faster responses. 18,23,25,27,28 With increasing treatment options in R/M SCCHN, sequencing becomes important. However, sequencing has not been addressed well in the clinical trials. In KN -048 study, only 16.7% patients in cetuximab arm received subsequent immunotherapy, while in IC or immunotherapy arm only 16.4% patients received subsequent EGFR inhibitor. 29 So, the right sequencing of therapies in RM SCCHN is still elusive.
Here we report our experience with 1 st -line weekly PCC combination followed by 2 nd -line immunotherapy (nivolumab) in platinum-sensitive R/M SCCHN from two tertiary care centers of India.

Methods
This study involved a retrospective analysis of consecutive patients with platinum-sensitive R/M SCCHN treated between August 2017 and May 2020 at the two tertiary care centres in India with 1 st -line weekly combination paclitaxel 80 mg/m 2 , carboplatin AUC2 and cetuximab 400 mg/m 2 loading followed by 250 mg/m 2 (PCC) for a maximum duration of 12 weeks. Non-progressive patients were then started on cetuximab maintenance. Patients were further treated with 2 nd -line nivolumab or oral metronomic chemotherapy (OMCT) on progression as per feasibility. Eligible patients 18-70 years old with ECOG-PS 0-2 and had at least one measurable lesion as per Response Evaluation Criteria in Solid Tumors version 1.1 (RECIST 1.1) were included. Exclusion criteria included any previous systemic chemotherapy for HNSCC surgery or radiotherapy within the previous 6 weeks; a previous dose of cisplatin more than 300 mg/m 2 ; treatment with EGFR-targeting therapy within the previous 12 months; clinically significant cardiovascular disease; other malignancies within 5 years before randomisation, Written consent was obtained from the patients.
Demographic, clinical and treatment characteristics data were collected from patients' charts until July 7, 2020. Response evaluation was performed as per RECIST 1.1 (for solid tumors) after every 8-12 weeks. 30 Toxicities during treatment were documented in accordance with CTCAE version 4.03. 31 The institutional ethics committees of the Narayana Superspecialist Hospital, Howrah and Dharamshala Narayana Superspecialist Hospital, New Delhi approved the study. Written informed consent was obtained from each patient.

Study objectives
To evaluate the overall response rates (ORR), progression-free survival, and overall survival (OS) after 1L PCC and 2 nd line therapies. The study also evaluated the safety of 1L PCC and second-line therapies.

Statistical analysis
Statistical measures were calculated using software SPSS Version 20. PFS and OS were estimated by Kaplan-Meier method and compared by log rank test. Median follow-up was calculated using the reverse Kaplan Meier method. Mean AE S.D. and Median were used to summarize the quantitative variables.

Baseline characteristics
Fifty-four patients with R/M SCCHN were treated at the two centers between August 2017 and May 2020. Median age of the study population was 56.5 years; 89% were males. Oral cancer was the primary cancer site in 30 (55%) patients and 24 (45%) had non-oral primary cancer sites. Of the 54 patients, 16 (30%) were treatment naïve; 38 (70%) had received prior treatments (chemotherapy or surgery or radiation or a combination of these) and had a median treatment free interval (TFI) of 10 months; 28 (51%) had not received any previous systemic therapy. A history of tobacco abuse was present in 41 (75%) of patients. All patients had Eastern Cooperative Oncology Group performance status (ECOG PS) of 1 or 2 at the start of PCC; 6 (11%) had cardiac dysfunction and 7 (13%) had renal dysfunction. Demographic details of the study population are provided in Table 1.

Treatment outcomes
The study schema is outlined in Figure 1. All patients received 1L weekly combination of paclitaxel 80 mg/m 2 , carboplatin AUC2 and cetuximab 400 mg/m 2 loading followed by 250 mg/m 2 (PCC) for up to maximum 12 weeks, followed by weekly cetuximab maintenance 250 mg/m 2 until disease progression or intolerable adverse effects. Thirtythree (61%) patients progressed on 1L PCC and 21 (39%) did not. The patients who progressed on 1L PCC, were further treated with second line (2L) nivolumab 3 mg/kg two weekly (n = 19) or oral metronomic chemotherapy (n = 10) (OMCTcelecoxib 200 mg twice daily + erlotinib 100 mg once daily + methotrexate 12 mg/m 2 once weekly) as per feasibility. Four patients who progressed on 1L PCC were managed on best supportive care (BSC) alone. Nonprogressive patients were given single-agent cetuximab in maintenance dose.
The median number of Cetuximab cycles used as 23. Outcomes of the study population in 1L PCC is given in Table 2 The best response after second line therapies is summarized in Table 3. The ORR was 26.3% with nivolumab and 10% with OMCT. Median PFS and OS after second line therapy were 6.5 months and 20.6 months with nivolumab, whereas 2 months and 7 months with OMCT, respectively (Figure 2).
After a median follow-up of 21 months, 30 patients were alive; 21 on 1L PCC who were on cetuximab maintenance and nine on 2L nivolumab; 24 succumbed to the disease.

Discussion
Current study shows that weekly paclitaxel-carboplatin appears as an equally effective backbone for cetuximab with acceptable safety profile as compared to EXTREME regimen (cisplatin-5FU) 16,17 or TPExtreme regimen (docetaxelplatinum) 18 (Table 3). 20 The TPExtreme regimen was more tolerable than EXTREME with lower grade ≥3 adverse events (34% versus 50%) and had more patients initiating maintenance cetuximab than EXTREME regimen (73% versus 53%). In our study also, the PCC regimen had an acceptable safety profile that was safer than EXTREME and comparable to TPExtreme.
Our results of efficacy and safety of 1 st line PCC are in tandem with other data on 1 st -line PCC like CSPOR-HN02 32 and CACTUX 33 study (Table 4). The CSPOR-HN02 study which used split dose paclitaxel (100 mg/m 2 on days 1 and 8 every three weeks) showed that the PCC regimen can be given on an outpatient basis, thereby reducing cost of treatment. 21  In our study also, PCC was given on outpatient basis. This is an important aspect to consider in a country like India as it significantly reduces cost of treatment and need for admission beds. However, the CSPOR-HN02study did not look at treatment after progression on 1 st line PCC. On the other hand, only 22% of patients in the CACTUX trial received PD-1 inhibitor immunotherapy after progression and demonstrated a: median PFS of 2.2 months. 33 Comparatively, in our study, 19 of 33 patients who progressed (57.5%) received immunotherapy (nivolumab) and showed good response with an ORR of 26.3% and median PFS and OS of 6.5 months and 20.6 months, respectively.
Though EXTREME regimen continues to be the 1 st -line standard of care in R/M SCCHN, many patients are ineligible for cisplatin-based regimen primarily due to renal and cardiac dysfunction. 34,35 We found that PCC was safe and well tolerated even in patients with cardiac and renal dysfunction. Patients in our study could receive median of 23 cycles of Cetuximab based therapy.
We do note that early immunotherapy has shown OS advantage in KN48 study 26 and that the PFS-2 of 11months reported is superior to that in our study. However, KN-48 data may not be applicable in patients of Indian subcontinent. Unlike Indian patients, who predominantly have oral cavity primaries and mainly tobacco related etiology, KN-48 patients had mainly non-oral primaries and HPV related etiology. Other barriers to the ease of using immunotherapy front line are the accessibility to PDL-1 combined positive score testing and its relatively higher cost. Also, the OS results in the Keynote -048 study are mainly driven by the PD-L1 combined positive score>20 subgroup.
Another matter of concern is that, more than 1/3 patients in the pembrolizumab monotherapy group may have frank disease progression in the initial three months of therapy. This is more worrisome in Indian patients because they have a far higher disease burden than that of the western population. Hence, clinicians may lose this narrow window of opportunity and the patients may not remain eligible for subsequent anticancer therapy in view of rapid disease progression and deteriorating performance status. Hence response rate become a more important endpoint in Indian patients and may act as a surrogate to better quality of life and survival. However, a randomized prospective trial comparing the alternate sequencing of immunotherapy-based and cetuximab-based therapy in R/M HNSCC will be able to better guide the efficacy, safety and sequencing of these drugs.
The study is limited by its retrospective design and small patient population. Additionally, PD-L1 expression was not performed, so we cannot access which patient population actually benefited from immunotherapy. Despite this our study has contributed immensely to the growing body of knowledge that cetuximab based PCC regimens can be used effectively and safely in 1 st -line palliative care of R/M SCCHN. To the best of our knowledge, this is the first real-world data of 1 st -line PCC followed by immunotherapy from Asia.

Conclusions
First-line weekly PCC is an effective regimen for palliative therapy of platinum-sensitive R/M SCCHN with acceptable toxicity profile. Addition of 2 nd -line nivolumab on progression, further improves the outcomes.