Carboplatin in combination with weekly Paclitaxel as first-line therapy in patients with recurrent/metastatic head and neck squamous cell carcinoma unfit to EXTREME schedule

The standard first-line treatment in recurrent/metastatic head and neck squamous cell carcinoma combines Cisplatin, 5 Fluorouracil and Cetuximab, but many patients aren’t eligible. We retrospectively evaluated the efficacy and the tolerability of Carboplatin and Paclitaxel in this indication, mostly in patients unfit to Cisplatin. Paclitaxel (80mg/m2) was administered at day 1, 8 and 15 and Carboplatin area under the curve 5 at day 1, repeated every 28 days, for 6 cycles. Carboplatin could be administered at area under the curve 2 at day 1, 8 and 15. 117 patients received this association at our institution, 94 of those were ineligible to cisplatin due to severe comorbidities, age >70years or Performance status >1. The overall response rate was 40%. The median progression free survival for patients ineligible to Cisplatin was 4.4 months [95% CI; 3.4; 5.0] and the median overall survival was 8 months [95% CI; 5.4–10.7]. The most frequent toxicities were hematologic, with 94 grade ≥ 3, mostly in patients who received monthly Carboplatin. Our study shows Carboplatin and Paclitaxel in first-line in recurrent/metastatic head and neck squamous cell carcinoma appear efficient for patients ineligible to Cisplatin and safe when both drugs are weekly administered.


INTRODUCTION
Head and neck cancer accounts for 500,000 new cases and nearly 300,000 deaths annually worldwide [1] .
Squamous cell carcinoma is the most frequent histological subtype of head and neck tumors [2].
About two thirds of head and neck squamous cell carcinoma patients are diagnosed with locally advanced disease [3] and are treated with a combination of surgery, radiation therapy and chemotherapy. Despite this primary treatment, more than a third of these patients have locoregional recurrences or distant metastases [4][5][6][7] and in this situation treatment is mostly palliative and disappointing, with a median overall survival (OS) of less than 1 year [8].
About 10% of the patients present with distant metastases at diagnosis [9]. The most frequent site of distant metastasis is the lung [3].
Platinum based chemotherapy is the usual firstline palliative treatment. Currently, the standard of care is the combination of Cetuximab, Cisplatin and 5-Fluorouracil, followed by maintenance with Cetuximab (the "EXTREME" regimen) [10] .
A large proportion of patients are ineligible to the EXTREME regimen because of their age>70yo, Performance Status (PS)>1 or their severe comorbidities (cardiac or renal insufficiency etc…). In this population, despite replacement of Cisplatin with Carboplatin in the EXTREME regimen, tolerability remains poor.

Research Paper www.oncotarget.com
In advanced non-small cell lung cancer, the standard of care is platinum based chemotherapy too. In elderly population, Carboplatin with weekly Paclitaxel is a safe and recommended option [11,12].
Similarly to this indication in lung cancer, we propose in our institution Carboplatin with weekly Paclitaxel for patients unfit to Cisplatin and in some patients eligible to Cisplatin who refuse the EXTREME regimen for various reasons (mostly due to the 4 days of continuous perfusion of Fluorouracil).
This association isn't validated in head and neck cancer but several studies have shown the efficacy of Paclitaxel as monotherapy or in combination with Cisplatin [13] with promising results.
In this retrospective study we evaluated the efficacy and the tolerability of Carboplatin and Paclitaxel as firstline treatment in patients with recurrent or metastatic head and neck squamous cell carcinoma (RMHNSCC), mostly for patients who are unfit to the EXTREME regimen.

Patients' characteristics
Between August 2009 and December 2016, 117 patients with RMHNSCC were treated at the "Centre Léon Bérard "(Lyon, France) with Carboplatin and Paclitaxel combination as first-line therapy.
Patients' characteristics are summarized in Table 1. Since 89% of patients experienced a recurrence, 41% received a platinum-agent previously for a localized disease (18% of those less than 6 months before relapse). Before palliative chemotherapy onset, 34 (29%) patients were treated locally for previous relapses (mainly surgery, radiotherapy potentiated or not by chemotherapy).The most common site of distant metastases was pulmonary .
All patients received Carboplatin and Paclitaxel strictly in first intent except for 8 patients. 2 of these patients should be retrospectively considered as initially metastatic: they had non-specific lung micronodules when they started induction chemotherapy with Cisplatin Docetaxel and Fluorouracil. After that they were treated with radiotherapy and one of them potentialized with Cisplatin. After 7.8 and 10 months of the last platin administration, the lung nodules progressed and they started carboplatin and paclitaxel. Despite a certain frailty, 5 patients received a first cycle of cisplatin-based chemotherapy and switched to Carboplatin and Paclitaxel due to severe toxicity. Finally 1 patient received a cycle of Cisplatin/Vinorelbine since the solitary metastasis was considered as a primary lung cancer and switched at the second cycle for Carboplatin with Paclitaxel.
Of the 117 patients treated with Carboplatin and Paclitaxel, 23 (20%) were eligible to a chemotherapy by Cisplatin Fluorouracil and Cetuximab (EXTREME regimen).
3 of those received firstly a Cisplatin-based chemotherapy, then, due to toxicity, Carboplatin and Paclitaxel. 94 (80%) were ineligible to EXTREME due at least to one adverse criteria among age >70 years, renal failure (creatinine clearance <60 ml/min), PS≥2 or severe comorbiditie(s) (cardiac insufficiency, cirrhosis, cisplatin allergy etc…). Details of ineligibility are summarized in Table 2.

Carboplatin and paclitaxel delivery
Data on chemotherapy delivery are summarized in Figure 1. Initially, Carboplatin was administered on a monthly basis for 88 (75%) patients and weekly for 29 (25%) patients. Due to severe frailty, 35 patients received a 25% reduction of the dose of a least one agent. Of note, 3 patients received Cetuximab in combination with carboplatin and paclitaxel. Treatment toxicities are presented in Table 3.
Dose reduction was needed for 60 (51%) patients, 11 of those resulted with a modification of monthly Carboplatin to weekly Carboplatin and 25 (21%) stopped treatment due to toxicity.
The most common side effects were hematologic, they are detailed in Table 4. There were 94 grade ≥ 3 toxicities with 17 (15%) febrile neutropenia, 2 of those were fatal (2%). 18 patients needed granulocyte colony stimulating factors.
Overall, weekly Carboplatin was better tolerated, inducing fewer severe toxicities than monthly Carboplatin.

DISCUSSION
The EXTREME (Platin-5FU-Cetuximab) regimen has become the standard of care in first-line therapy in patients with recurrent/metastatic head and neck squamous cell carcinoma with 36% responses and a median overall survival of 10.1 months [95% CI; 8.6-11.2] (versus 20% and 7.4 months [95% CI; 6,4-8,3] respectively for Platin-5FU) [10]. In subgroup analysis, OS was significantly better only with cisplatin at 10.6 months (vs 7.3 with chemotherapy alone) whereas there was not any significant difference with carboplatin (9.7 vs 8.3 months) suggesting Again, even if a comparison isn't possible between the results of a clinical trial and our retrospective study, Carboplatin and weekly Paclitaxel seems better for patients ineligible to Cisplatin with PS 0/1 compared to Carboplatin+5 Fluorouracil+Cetuximab in EXTREME study with a median OS of 11.5 months [95% CI; 8.5-19.1] vs 9.7 months and a median PFS of 5.5 months [CI 95%; 4.8; 7.1] vs 5.3 months. Survival was poor for patients with PS2 or 3 with OS of 3.6 and 1.7 months and PFS of 2.0 and 1.7 months respectively. Since this population is usually excluded from clinical trials we cannot draw definitive conclusion. Specific studies would be necessary to evaluate this population. Furthermore our definition of  PFS was different than EXTEREME study, we considered the deaths only from cancer and not from any cause, it could be overestimated our results. In our study, the tolerance of Carboplatin and weekly Paclitaxel appeared acceptable in this frail population with weekly Carboplatin regimen. In fact, the toxicities with monthly Carboplatin were more frequent and more severe. And a large part of the population of weekly Carboplatin group switched from monthly group after toxicities. Response rate of weekly Carboplatin group delivered from the beginning was slightly lower than for the overall population at 34% but we cannot conclude with   In further studies we propose to adopt a weekly regimen for both Paclitaxel and Carboplatin. The main toxicities were hematologic with 26% and 3% anemia grade ≥3, 8% and 0% thrombopenia grade ≥3 and 40% and 28% neutropenia grade ≥3 in monthly and weekly Carboplatin group respectively.
Hepatotoxicity with Carboplatin and Paclitaxel is described in several studies [11,14]. In our population, we didn't observed any hepatic perturbation related to chemotherapy.
Toxicities of the Carboplatin group of the Extreme study are not available but in our practice it seems to be less toxic than Cisplatin, so we cannot formally compare our data.
Another study demonstrated safety and efficacy of weekly Carboplatin and Paclitaxel in 31 patients with locally advanced, distant metastases or recurrent head and neck squamous cell carcinoma. Median OS was 12.8 months [95% CI 8.6-15.5] and the major toxicity was hematologic too, with 22% neutropenia grade ≥3, 12% anemia grade ≥3 and 0% thrombopenia grade ≥3 [15].
With our results and since monotherapies with Paclitaxel, Cetuximab or Capecitabine are efficient after failure of platinum [16][17][18] there is sense to prefer bichemotherapy (by Carboplatin and Paclitaxel) to trichemotherapy (by Platin-5FU-Cetuximab) in order to allow sequential treatments that could increase survival [19]. Moreover, previous studies showed that Cetuximab could be combined with Paclitaxel with possible synergy and promising efficacy after failure of Platinum [6,20] and thus could be better than EXTREME schedule for selected patients [21]. So, combination of the three agents Platin, Paclitaxel and Cetuximab could be of great interest. Indeed, a randomized phase II study in first-line reported 51.7% responses with Cisplatin-Paclitaxel-Cetuximab and an OS of 11 months [22]. Similarly, another phase II study combining Cisplatin Docetaxel and Cetuximab in first-line showed 44.4% responses and OS of 14 months [23].
In conclusion, our study shows that weekly Carboplatin and Paclitaxel is a good option as first-line therapy for recurrent/metastatic head and neck squamous cell carcinoma, mostly for patients ineligible to Cisplatin. A future question would be the place of Cetuximab: in combination with Carboplatin and Paclitaxel or after failure of Nivolumab (or another anti-PD1 or PDL1). Similarly, combination of Carboplatin and Paclitaxel with immunotherapies could be relevant as in advanced nonsmall cell lung cancer [27,28].

Patient selection
We retrospectively reviewed the data in our institution between August 2009 and December 2016.
The inclusion criteria were as follows: (1) 18 or older patients with patients with histologically confirmed head and neck squamous cell carcinoma, (2) treated with Carboplatin and Paclitaxel in first-line for a recurrent or metastatic disease.
A first-line is defined by a first treatment in the recurrent/metastatic setting or by a switch from another treatment (due to toxicity of the first course) before any progression.
Prior treatment for a localized disease by surgery, radiotherapy or chemotherapy were permitted .

Treatment
Different 28-days schedules were used. Paclitaxel was administered at 80 mg/m2 at day 1 (D1), day 8 (D8) and day 15 (D15). The standard schedule was Carboplatin AUC5 (area under the curve = 5) at D1 and in the alternative schedule Carboplatin was administered at AUC2 at D1, D8 and D15.
The doses could be reduced at the beginning or during treatment according to the patient's frailty or to toxicities.

Assessments
Response was evaluated every 6-8 weeks by repeated clinical and computed tomographic scan assessments on the basis of the extent of disease at presentation. Antitumor activity was evaluated according to the Response Evaluation Criteria In Solid Tumors criteria 1.1 [29].

Statistical design
Overall survival (OS) was defined as the time from the date of first Carboplatin and Paclitaxel administration to the date of death.
Progression-free survival (PFS) was calculated from the date of first Carboplatin and Paclitaxel administration to the date of progression or death secondary to the cancer, whichever occurred first. If progression or death did not occur before the cut-off date, data were censored at the time of the last valid assessment.
Survival distributions were estimated by the Kaplan-Meier method.

Author contributions
AP performed data analysis, wrote the manuscript. AK edited the manuscript. EVN, PER, MP, SD and PZ contributed to data acquisition. JF supervised data acquisition and analysis, wrote and edited the manuscript.

CONFLICTS OF INTEREST
Jérome Fayette has been paid honoraria by BMS.

FUNDING
This study was not supported by any pharmaceutical corporation or company.