Systemic Therapy in Recurrent or Metastatic Squamous Cell Head and Neck Cancer

Most of SCHNC are locally advanced at diagnosis and are treated with combination of surgery and/or radiotherapy (RT) with chemotherapy (CHT).[-] In spite of aggressive treatments almost half of these patients relapse within 3 years of initial diagnosis. Those whose tumor is unresectable or cannot be reirradiated are treated with a systemic treatment mostly in a palliative setting. They are identified as recurrent and/or metastatic SCHNC (R/M-SCHNC) patients.

The treatment of unresectable R/M-SCHNC is generally dictated by patient's performance status (PS) and intent to treatment (palliative vs. curative). A vast majority of these patients, however, have unresectable disease and only qualify for palliative treatment with systemic therapy. First-line treatment of R/M-SCHNC historically consisted of cytotoxic agents such as Methotrexate (MTX), Bleomycin, or platinum-based protocols until targeted biological therapies were introduced in the 2000's.[] Despite considerable progress made in the use of CHT, RT, and targeted therapies, in majority of patients with R/M SCHNC the survival outcomes remain poor.[]

The recent years witnessed a shift in systemic treatment toward the use of monoclonal antibodies and tyrosine kinase inhibitors, largely based on recent understanding of the role of immune dysfunction in SCHNC, including the observation that antagonizing the programmed cell death (PD-1) immune checkpoint can disable T-cell suppression by R/M-SCHNC cells for re-sensitization of the immune system to clear tumor cells.[]

Indeed, HNC, like all other cancers, is a result of a stepwise accumulation of genomic instability, chromosomal aberrations, and genetic mutations,[] with common features of HNCs also including tumor-mediated inhibition of antitumor immune responses and a high mutational burden. While the number of promising immune-based therapies continuously rises, efficiency of these is expected to rapidly improve with the possibility of patients" selection based on personal immunogenomic profiles. The emerging role of immunotherapy (IMT) as a potentially beneficial addition to standard treatments for R/M-SCHNC offers hope to the patients for whom no other therapeutic options exist.[]

However, despite the optimism, the optimal regime is still unknown. Many recommendations are based on a single randomized controlled trial (RCT) results with current lack of efficacious treatment options available when FDA-approved 1^st and 2^nd line treatments are contraindicated.[] The present manuscript reviews recent developments of molecular-targeted and immunotherapies in the treatment SCHNC, focusing on R/M-SCHNC. It also highlights ongoing trials and discusses some promising novel targets in HNC, as well as clinical trial design challenges.

First-line Treatment
In locally advanced but M0 SCHNC concurrent administration of RT and CHT is a standard of care in both HPV? and HPV+HNC.[-] Cisplatin (CDDP) is mostly used single-agent with radical RT (mostly 70 Gy in 35 daily fractions over 7 weeks) administered as three cycles of 100 mg/m² on days 1, 22, and 43 of the RT course. Due to a significant toxicity and frequent dose reduction and treatment interruptions of this approach, researchers attempted to offer an alternative schedule using mostly 40 mg/m2 given weekly for 7 weeks. This was based on the assumption that it would lead to a lesser toxicity while, at the same time, offering better (more prolonged) radio sensitization due to a more frequent CDDP administration.[,] In HPV+SCHNC, majority occurring in oropharynx, current trends in clinical research focus on de-intensification of treatment approach as data shows higher radiosensitivity of HPV+SCHNC.[]

Chemotherapy (CHT)
Various trials compared the activity of single cytotoxic agents. Response rates to single agent therapies range from 15% to 35%.[-] MTX was compared to taxane therapy and found objective response rate (ORR) favoring taxanes (OR=3.16, 95% CI: 1.26-7.97 p=0.01), without the difference in the overall survival (OS). Exception to this was increased progression-free survival (PFS) when Paclitaxel was given over 24 h but at the expense of higher toxicity. No other single agent increased ORR, PFS, and OS.[-] Two RCTs demonstrated a superior ORR of CDDP-5-Fluorouracil (5- FU) when compared to CDDP alone (OR=2.44, 95% CI: 1.50-3.95, p<0.0003), which did not lead to OS advantage for the combination regimen.[,]

Targeted Therapy (Epidermal Growth Factor Receptor [EGFR], EGFR Combination Therapy, PI3K Inhibitors, and VEGF Inhibitor)
EGFR expression is associated with poor prognosis and resistance to therapy and it occurs in up to 90% of SCHNC.[] Vermorken et al.[] tested single agent Cetuximab in a phase II trial enrolling 103 patients with R/M SCHNC who had progressed on platinumbased CHT. The ORR for Cetuximab alone was 13% and the stable disease rate was 33%. Median time to progression and median survival time (MST) were 2.3 and 6 months, respectively.

The phase III IMEX study compared an anti-EGFR agent Gefitinib with MTX in patients who had at least one prior therapy for R/M SCHNC and showed that Gefitinib was not superior to MTX in this setting.[]

Combination therapy of Cetuximab, CDDP and 5-FU (the EXTREME regimen) became the standard for first-line treatment in 2008 based on its OS benefit when compared to CHT alone in this phase III RCT.[] Cetuximab with platinum-based CHT significantly prolonged the MST from 7.4 months in the CHT-alone group to 10.1 months in CHT/Cetuximab group (HR 0.80; 95% CI:, 0.64 to 0.99; p=0.04) as well as it prolonged median PFS (MPFS) from 3.3 to 5.6 months (HR for progression, 0.54; p<0.001).[]

In multicenter open-label, randomized, phase 2 trial, six cycles of EXTREME regimen was compared to four cycles of TPEx regimen (CDDP, docetaxel, and cetuximab). In both arms in case of disease control after four cycles, cetuximab was continued as maintenance therapy until progression or unacceptable toxicity. Although the trial did not meet its primary endpoint (OS), with no significant improvement in OS with TPEx versus EXTREME, the TPEx regimen had a favorable safety profile. The TPEx regimen could provide an alternative to standard of care with the EXTREME regimen in the first-line treatment of patients with R/M HNSCC, especially for those who might not be good candidates for up-front pembrolizumab treatment.]

Panitumumab with platinum-combination shows increasing of ORR and PFS (HR 0·780, 95% CI: 0·659-0·922; p=0.0036) without prolongation of OS (HR=0.87, 95% CI: 0.72-1.04, p=0.1403).[]

The addition of Bevacizumab to platinum-doublet CHT did not improve OS but improved the ORR and PFS with increased toxicities.[]

Immunotherapy (IMT)
While the concept of the immune system being able to recognize and control cells undergoing malignant transformation exists for more than a century,[] recent work focused on understanding the immunobiology of SCHNC and, consequently, on developing strategies to promote an antitumor immune response. [] Introduction of immune checkpoint inhibitors (ICI) has significantly changed the therapeutic approach in HNC. SCHNC shows a high mutational burden and also present with a high infiltration of immune cells, demonstrating that they are immunogenic through the expression of epitopes, virus or not virusrelated. While this may have an important implication in the future; however, the most important current limitation of potential biomarkers is the impossibility to identify responding patients. As a consequence, TMB testing in HNSCC is not yet recommended as a standard of care.[]

For R/M-SCHNC, tumor programmed death-ligand 1 (PD-L1) expression should be evaluated and PD-L1 expression is assessed by two methods: The tumor proportion score (TPS), defined as the percentage of tumor cells with membranous PD-L1 staining, or by the combined positive score (CPS), defined as the number of PD-L1-positive cells (tumor cells, lymphocytes, and macrophages) divided by the total number of tumor cells multiplied by 100. The CPS can help to define the first-line treatment strategy for R/M-SCHNC.[]

Pembrolizumab
In the Phase Ib Keynote-012 study, both treatment- naive and pretreated patients (n=60) with R/M SCHNC with ≥1% of PD-L1 expression were treated with Pembrolizumab. Among the heavily pretreated patients 70% were previously treated with ≥2 lines of CHT for R/M disease. Several non-progressing patients continued therapy beyond progression. Of these, 18% demonstrated an ORR while 51% experienced any reduction in tumor burden. Median time to response (MTTR) was 8 weeks and the median duration of response (MDR) was 53 weeks. MPFS and MST were 2 and 13 months, respectively. HPV status did not affect the effect of Pembrolizumab. PD-L1 expression levels were associated with ORR and PFS.[] A larger expansion cohort of the Keynote-012 trial,[] with 132 patients with unrestricted PD-L1 positivity level, confirmed the results with ORR of 18%, the MPFS and MST of 2 and 8 months, respectively. Response to Pembrolizumab was again unrelated to HPV status.

In the subsequent nonrandomized Phase II study (Keynote-055),[] 171 patients with R/M SCHNC resistant to both platinum and Cetuximab were treated with pembrolizumab 200 mg/3 weeks. ORR was 16% with a MTTR of 2 months, and MDR of 8 months. MPFS and MST were 2.1 months and 8 months, respectively, with no influence of HPV status on clinical activity.

In a randomized phase 3 study (KEYNOTE-048), 882 patients with untreated locally incurable R/MSCHNC were stratified by PS, p16 status, and PD-L1 expression and randomly allocated (1:1:1) to Pembrolizumab alone, Pembrolizumab plus a platinum and 5-FU (Pembrolizumab with CHT), or Cetuximab plus a platinum and 5-FU (Cetuximab with CHT). Of these, 754 (85%) had CPS of ≥1 and 381 (43%) had CPS of ?20. At the final analysis, Pembrolizumab with CHT improved OS versus Cetuximab with CHT in both total population (13.0 months vs. 10.7 months, HR 0.77 [95% CI: 0.63-0.93], p=0.0034), in the CPS of ≥20 population (14.7 vs. 11.0, HR, 0.60 [95% CI: 0.45-0.82]; p=0.0004) as well as in the CPS of ≥1 population (13.6 vs. 10.4, HR, 0·65; [95% CI: 0.53-0.80], p<0.0001). Neither Pembrolizumab alone nor Pembrolizumab with CHT improved PFS at the second interim analysis. Grade ≥3 all-cause adverse events were less frequent (55%) in the Pembrolizumab alone group, than in the Pembrolizumab with CHT group (85%), or Cetuximab with CHT group (83%). Based on the observed efficacy and safety, Pembrolizumab with CHT seemed an appropriate first-line treatment for R/M-SCHNC while single agent Pembrolizumab seemed an appropriate first-line treatment for PD-L1-positive R/M-SCHNC.[]

However, in a subgroup analysis, no survival benefit was found in patients presenting with local or regional recurrence only, irrespective of the CPS or assigned IMT arm (monotherapy or combination treatment), indicating that patients with a substantial locoregional disease burden represent a particularly challenging group. Indeed, the rates of progressive disease in patients receiving Pembrolizumab alone as first-line treatment in KEYNOTE-048 was quite disappointing (32-41%). Although being substantially lower for Pembrolizumab plus CHT (15-17%), they still remain numerically higher than in the CHT arm (8-12%) irrespective of CPS. Furthermore, some patients might experience an accelerated tumor growth known as hyperprogression. Caution must be advocated when treating patients with high locoregional disease burden, in particular when it concerns single agent Pembrolizumab. Here, the risk of progression should be weighed against the reduced morbidity with IMT and warrants individual decision making.[] Finally, current evidence does not point to platinum/5-FU/Pembrolizumab improving survival compared with platinum/5-FU/Cetuximab in patients with SCHNC not expressing PD-L1. The impact of Pembrolizumab on survival in patients with SCHNC and a CPS between 1 and 19 also calls for further investigation.

The phase 3 randomized, KEYNOTE-040 trial, in which patients who failed prior platinum-based CHT failed were included, compared Pembrolizumab 200 mg/3 weeks to the drug of investigator choice. Randomization ratio was 1:1 and stratification by ECOG PS (0 vs. 1), p16 status for oropharyngeal tumors (positive vs. negative) and PD-L1 TPS (TPS >50 vs. <50%). The primary endpoint, the MST, was 8.4 months in the Pembrolizumab arm versus 6.9 months in the investigator choice arm (95% CI: 0.65-0.98; nominal p=0.0161). In the subgroup of patients with a TPS ≥50%, the MST was 11.6 months with Pembrolizumab and 6.6 months with standard of care (HR 0.53, p=0.0014). In the subgroup of patients with TPS <50%, the MST was 6.5 months with Pembrolizumab and 7.1 months with standard of care. Grade ?3 toxicity was rare (13%).[]

Nivolumab
In randomized phase III (CHECKMATE-141) study[] 361 SCHNC patients with tumor progression or recurrence within 6 months after the last dose of platinum-containing CHT were treated with Nivolumab 3 mg/kg every 3 weeks versus investigator's single agent CHT choice (MTX, Docetaxel or Cetuximab). The MST was 7.5 months in the Nivolumab group versus 5.1 months in the control group, with no difference in PFS (2 vs. 2.3 months). The ORR and duration of response were higher for Nivolumab-treated patients (13.3 vs. 5.8%). The safety profile was also in favor of the Nivolumab with less frequent grade 3-4 AEs (13.1 vs. 35.1%) as well as the quality of life.[] A 2-year update confirmed the benefit for survival (2-year OS rate: 16.9 vs. 6.0%) and safety (grade 3-4 AEs: 15.3 vs. 36.9%) of Nivolumab on investigator choice.[]

Durvalumab
In the 1108 study, 62 pretreated R/M SCHNC patients received Durvalumab monotherapy, obtaining seven responses, with six patients showing response duration of >12 months. MPFS and MST were 1.4 months and 8.4 months, respectively. OS was 62% at 6 months and 38% at 12 months (42% for PD-L1 ≥25%, 36% for <25%).[]

The single-arm, phase II HAWK study evaluated Durvalumab monotherapy, in IMT -naïve patients with platinum-refractory R/M SCHNC having high PD-L1 (TC≥25%). Among 111 evaluable patients, ORR was 16.2% (95% CI: 9.9-24.4). MPFS and MST were 2.1 months and 7.1 months, respectively, while PFS and OS at 12 months were 14.6% (95% CI: 8.5-22.1) and 33.6% (95% CI: 24.8-42.7), respectively. Grade ≥3 AEs were 8.0%, and none led to death. These results supported its ongoing evaluation in phase III trials in first- and second-line setting. In an ad hoc analysis, HPV-positive patients had a numerically higher response rate and survival than HPV-negative patients.[]

Atezolizumab
Another anti-PD-L1 agent, Atezolizumab showed in all patients, regardless of PD-L1 expression, the confirmed ORR of 22%, MPFS of 2.6 months and MST of 6.0 months, seemingly comparable to the results other ICIs achieved. In this phase I trial, after the first ten patients were non-selectively enrolled, identification of PD-L1 as a potential biomarker led to subsequent enrolment based on PD-L1 status of >5% expression on immune cells. Among 32 treated patients, a slightly higher ORR was observed in case of high PD-L1 expression (24%) than in those with low or no expression of PD-L1 (14%).[]

Combination IMT
The phase 3 EAGLE trial compared Durvalumab monotherapy (A) or Durvalumab in combination with Tremelimumab (B), an anti-CTLA4 antibody versus standard CHT (C) regardless of the PD-L1 status. No statistically significant improvements in OS were observed for A versus C (HR: 0.88; 95% CI: 0.72-1.08; p=0.20) or B versus C (HR: 1.04; 95% CI: 0.85-1.26; p=0.76). One-year survivals were 37.0%, 30.4%, and 30.5% for A, B, and C, respectively. Grade ?3 AE rates were 10.1%, 16.3%, and 24.2% for A, B, and C, respectively. In spite of no significant difference in OS for A or B versus C, authors suggested clinical activity for Durvalumab due to its higher survival rates at 12-24 months and higher response rates.[]

A phase 2, randomized, open-label study (CONDOR) included patients with disease progression or recurrence during or after treatment with only 1 platinum containing regimen for R/M disease. Patients were stratified by HPV and smoking status and then randomized 2:1:1 to treatment with (A) Durvalumab and Tremelimumab for four cycles, followed by Durvalumab, (B) Durvalumab monotherapy, or (C) Tremelimumab monotherapy for up to 12 months. ORR for the three groups was 7.8%, 9.2%, and 1.6%, respectively, while the MST for the three groups was 7.6 (95% CI: 4.9-10.6), 6.0 (95% CI: 4.0-11.3), and 5.5 (95% CI: 3.9-7.0) months, respectively. This study showed that Durvalumab monotherapy had a manageable toxicity profile and clinical benefit for patients with R/M SCHNC and low or no PD-L1 TC expression, while Durvalumab + Tremelimumab demonstrated similar efficacy to Durvalumab monotherapy.[]

According to updated efficacy and safety findings from a Phase II study dual CTLA-4/PD-1 blockade with Ipilimumab plus Nivolumab provided durable responses in 40 patients with R/M nasopharyngeal carcinoma (NPC) who have received no more than one prior line of CHT. Nivolumab was given at 3 mg/ kg/ 2 weeks and ipilimumab at 1 mg/kg/6 weeks. The best overall response was partial response, achieved in 14 (35%) patients (95% CI: 20.6-51.7%). Responding patients showed a MDR of 5.9 months (95% CI: 3.95- 8.97), MPFS of 5.3 months (95% CI: 3.0-6.4 months), and MST 17.6 months (95% CI: 13.1-30.0). Treatment- related AEs occurred in 34 (85%) patients and four (10%) patients had grade 3/4 serious AEs including hypocortisolism, pneumonia, myasthenia gravis, and increased lipase. No relationship was observed between response and either tumor mutation burden or PD-1 expression. This combination treatment was proven active based on durable responses and PFS data as well as safe in patients with NPC.[]

Toll-like Receptor (TLR) Agonists, Other Agents and Vaccines
Therapeutic antibodies against the TLRs are immunomodulatory oligonucleotides with an agonistic effect. The TLR-9 agonist EMD1201081 has been compared with Cetuximab in a randomized Phase II trial involving R/M SCHNC patients after failure of one CHT regimen. It showed a good tolerance but no therapeutic improvement over Cetuximab.[]

The preliminary results of a phase I/II study with 13 SCHNC patients receiving the TLR-8 agonist Motolimod showed disease control rate in 54% of patients when Motolimod was combined with Cetuximab.[] Although Motolimod added to the EXTREME regimen did not offer an improvement in either PFS or OS it provided significant improvement in both PFS and OS in patients with HPV positive oropharyngeal cancer.[]

When Paclitaxel was combined with PI3K inhibitor Buparlisib modest but promising response rates were observed.[] Other agents from the same category demonstrated different activity: While mTOR inhibitor Everolimus failed in two consecutive phase II trials,[,] Temsirolimus showed meaningful efficacy in another Phase II trial.[]

Due to ability of oncolytic therapy to selectively replicate in tumor cells causing direct cytotoxicity and inducing specific immune response against the tumor, several oncolytic viruses were tested in clinical trials. They have demonstrated safety and potential local tumor control in SCHNC due to both facility of intratumoral injection and importance of locoregional control in this disease. Current trials are ongoing in combination with PD-1 inhibitors.[]

Second-line Treatment (Targeted Therapy, IMT, TLR-Agonists)
Unfortunately, many patients with R/M SCHNC further relapse despite treatment. In second-line treatments no consensus exists on the optimal therapies. Participation in clinical trials represents preferred and recommended t option in many situations.

The choice of second line therapy was initially poorly defined until PD-1 inhibitors Nivolumab and Pembrolizumab became FDA- licensed for second-line treatment of R/M SCHNC for patients who had disease progression on or after platinum-based therapy.[,]

In second-line therapy none of single CHT agents demonstrated a superior clinical benefit over their control arms.[-] Cetuximab at escalating doses did not impact ORR, PFS, and OS.[] Cixutuxumab with or without Cetuximab demonstrated limited benefit to PFS and OS despite the increased ORR observed with Cixutuxumab and Cetuximab combination.[] When Gefitinib was added to Docetaxel, only a limited activity without improvements in ORR, PFS, and OS was observed.[]

Two studies assessed PI3K inhibitors.[,] PX- 866 added to Cetuximab failed to improve ORR, PFS, and OS over Cetuximab alone. Although ORR reached borderline significance (OR=3.88, 95% CI: 1.91-7.86, p=0.05), when PI3K inhibitor was added to taxane therapy,[50,64] PFS (HR=0.74, 95% CI: 0.55- 1.00, p=0.183), and OS (HR=0.86, 95% CI: 0.69-1.06, p=0.16) did not improve.

In the network meta-analysis by Jin et al.,[] 12 trials including a total of 10 and 12 second-line treatments were available for PFS and OS analysis, respectively. Nivolumab was the highest-ranked treatment for prolonging OS (0.95), while Buparlisib plus Paclitaxel was the highest-ranked treatment for PFS (0.94). This identified Nivolumab as the treatment of choice for overall R/M SCHNC patients due to its most remarkable OS benefit (HR 0.68, 95% CI: 0.58-0.80) and lower AEs ?grade 3 (OR 0.37, 95% CI: 0.11-1.22). Nivolumab was also significantly associated with improvement of OS in patients with high PD-L1 expression (HR 0.55, 0.43-0.70), but was similar to conventional CHT for those with low PD-L1 expression. Buparlisib plus Paclitaxel showed the best OS benefit in subgroups of patients with HPV-negative status.[]

New Targets and Ongoing Trials
The frequent RAS-RAF family pathway activation[] in SCHNC has made it an appealing conceptual target, although mostly unsuccessful in clinical scenario until recently. The RAS family is composed of three distinct genes (HRAS, KRAS, and NRAS), which can be pathologically activated by a series of mutations, some of whose sites are shared between the genes. HRAS is mutated in approximately 6% of SCHNC cases and overexpressed in higher proportions.[]

Ho et al.[] described promising results of a single- arm, open-label Phase II study when treating HRAS-mutated R/M SCCHN using the farnesyltransferase (FT) inhibitor Tipifarnib, which prevents FT from prenylating the HRAS protein CAAX tail motif. Inhibiting this prenylation prevents HRAS membrane binding and thereby renders it inactive. Of the 22 patients with variant allele frequency (VAF) >20% (high VAF), 20 were evaluable for response. Objective RR for evaluable patients with high-VAF SCCHN was 55% (95% CI: 31.5-76.9). MPFS on Tipifarnib was 5.6 months (95% CI: 3.6-16.4) versus 3.6 months (95% CI: 1.3-5.2) on last prior therapy. The MST was 15.4 months (95% CI: 7.0-29.7). The most frequent treatment- emergent AEs were anemia (37%) and lymphopenia (13%). The safety profile of Tipifarnib was tolerable and manageable in this Phase II trial. The study population had significant prior treatment with a median of two prior lines of therapy and the majority of patients had received IMT. These results represent potentially impactful change in clinical SCHNC research and successful targeting of HRAS would mark a significant step forward for precision oncology care in SCHNC.[] Unfortunately, a lot of patients with R/M SCHNC have no response to IMT, or initial responses are followed by disease progression. To improve response rates and survival, ongoing trials are evaluating various combinations of novel drugs including ICIs, therapeutic vaccines, and cytotoxic agents. Selected ongoing studies continue recruitment and further results are eagerly awaited (Table 1).

Table 1: Ongoing clinical trials in SCHNC