36.2.1 Molecular Markers of Prognosis

Analysis of surgical margins has yielded some potential prognostic markers relevant to locoregional recurrence risk. Nathan et al hypothesized that mammalian target of rapamycin (mTOR) signaling was relevant to the natural history of squamous cell carcinoma of the head and neck (SCCHN) and found that eIF4E-binding protein-1 (4E-BP1), which inhibits cap dependent translation, is overexpressed in a subset of head and neck surgical margins, and is associated with mTOR/Akt signaling. ¹³ This prompted the study of mTOR inhibition in FaDu cell lines of head and neck squamous cell carcinoma (HNSCC), and xenografts including a minimal residual disease model in which there was significantly less tumor growth among the cohort that was treated with the mTOR inhibitor CCI-779. ¹⁴

In another analysis, investigators from Johns Hopkins looked at the disrupted p53 status of resected head and neck tumors. TP53 mutations were detected in 53% of tumors; of these, 40% of the total patients were found to have disruptive mutations. In a multivariate analysis, there was an inferior survival among the patients with the disruptive TP53 mutation (HR = 1.7, p = 0.003). ¹⁵ This data has prompted the ongoing ECOG-ACRIN trial (EA3132, NCT02734537) in which patients with resected head and neck tumors, who lack the risk factors of + margin status and + ENE, are stratified by p53 status and randomized to radiation monotherapy versus cisplatin and radiation. ¹⁵

Given the known critical role of DNA damage and repair with regards to platinum and radiation sensitivity, biomarkers related to DNA repair have been another relevant area of study. ERCC1, which is a key component of the nucleotide repair pathway that typically repairs cisplatin-induced DNA adduct formation, has been assayed in head and neck cancers with immunohistochemistry (IHC) techniques. Specimens from 107 patients with SCCHN who had received cisplatin-based induction therapy were analyzed by IHC for ERCC1 levels; patients with low ERCC1 level had a four-fold greater odds of achieving an objective response to platinum-based chemotherapy. ¹⁶ A series published from Fox Chase Cancer Center, which consisted of surgically resected tumors from patients who then received adjuvant radiation or chemoradiation, utilized the ERCC1-specific antibody HPA029773. A longer median survival was observed in ERCC1-low versus ERCC1-high tumors (64 vs. 29 months; p = 0.02). ¹⁷

Additional biomarkers were studied in this human papilloma virus (HPV)-negative series, including the KRAS variant, a single nucleotide polymorphism in let-7 complementary site 6 (LCS-6) of the KRAS 3’UTR, which has been associated with platinum resistance, ¹⁸ and phosphorylated Rb. Among 99 available tumors cases from the Fox Chase series, 27% were noted to have the KRAS variant alteration. There was a non-significant trend toward increased overall survival among the KRAS-variant treated with surgery and adjuvant radiation (p = 0.08). ¹⁹

It is now known that dysregulation of the cell cycle pathway is of relevance in the pathogenesis of HPV-negative head and neck cancers. ²⁰ In The Cancer Genome Atlas (TCGA) analysis, amplification of cyclin D1 was identified and associated with a worse prognosis. ^{21 , 22} In order to study these pathways further, expression of epidermal growth factor receptor (EGFR) and inactivated phospho-T356 Rb was also quantified in 99 patients from the Fox Chase TMA. An improvement in survival was noted among the patients with low phospho-Rb. ²¹ In addition, an inverse relationship was noted between phospho-Rb and EGFR expression. ²³ While these biomarkers were associated with survival among HPV-negative patients who were treated with adjuvant radiation/chemoradiation, incorporation of these markers into therapeutic trials remains to be a question for the future.

36.3.1 Platinum-based Chemotherapy

The use of cytotoxic therapy, in particular cisplatin, had been incorporated with radiation in the setting of definitive treatment for curative intent. The mechanism of action of platinum involves induction of DNA damage via platinum-DNA adduct formation, resulting in inhibition of DNA replication and increased apoptosis. Preclinical data has demonstrated synergy with radiation, which results in single and double stranded DNA breaks, and cisplatin. This process is dependent on the presence of a functional nonhomologous end joining (NHEJ) DNA repair pathway which typically is involved in the repair of double strand breaks. ²⁴

For definitive treatment of SCCHN, results from the Radiation Therapy Oncology Group (RTOG) 91-11 trial demonstrated high rates of acute severe (grade 3 and 4) adverse events, specifically 47% hematologic side effects, 43% stomatitis, 35% dysphagia or odynophagia, and 20% nausea or vomiting. ²⁵ Only 120 of 172 patients (70%) treated with this regimen completed all three cycles of cisplatin. In the RTOG 0129 trial, once-daily fractionation RT for 7 weeks, in combination with three cycles of cisplatin (100 mg/m² on days 1, 22, and 43), was compared with accelerated boost RT (42 fractions given over 6 weeks) with two cycles of cisplatin (100 mg/m² on days 1 and 22). There was no statistically significant difference in overall survival when patients receiving two cycles of cisplatin with once-daily fractionation were compared with those receiving three cycles plus once-daily fractionation, although only a small number of patients received two cycles. ²⁶

Concurrent chemoradiotherapy with a platinum-based agent, such as cisplatin, has also become the standard for adjuvant chemotherapy. High-dose bolus cisplatin (100 mg/m² every 21 days) is often considered the preferred concurrent regimen, although it is associated with severe acute and late toxicities. As such, it is typically reserved for patients with excellent performance status and minimal comorbidities. ²⁷ Two randomized trials have established the efficacy of cisplatin with radiation in the adjuvant setting. The European Organization Research and Treatment of Cancer (EORTC) 22931 study randomized patients to radiation alone versus bolus dosing cisplatin and radiation. ²⁸ Eligibility criteria included pT3/pT4 stage with positive nodes, or an early stage tumor with N2/3 disease, extranodal spread, positive resection margins, or perineural involvement. The combined modality therapy with chemotherapy resulted in a statistically significant improvement in progression-free survival (PFS), LRC, and overall survival (HR = 0.70, p = 0.02). A similar study conducted in North America, RTOG 9501, included patients with high-risk features of two or more lymph nodes, ENE of nodal disease, and microscopically involved margins. Disease-free survival (DFS) favored the combined modality group, but there was not a significant benefit in overall survival (HR = 0.84, p = 0.19). ²⁹ A meta-analysis of these two studies confirmed the benefit of adjuvant chemoradiation among patients who had either ENE or positive margins. ³⁰

Alternative cisplatin dosing schedules (30-40 mg/m² weekly) are sometimes used because of improved patient tolerance. However, data comparing high-dose cisplatin with alternative regimens is limited. In a randomized phase III trial which included 87% oral cavity cancers, cisplatin, 30 mg/m² given weekly, resulted in significantly worse LRC compared with cisplatin 100 mg/m² every 3 weeks (2-year control rate 58.5 vs. 73.1%, HR = 1.76, 95% CI: 1.11-2.79). ³¹ However, there was no significant difference in overall survival with 22-month follow-up (median 39.5 vs. not reached, HR = 1.14, 95% CI: 0.79-1.65). This trial had several limitations and used a lower dose of cisplatin than is generally used (35-40 mg/m² weekly).

In another large phase III trial of patients in India with locally advanced oral cavity cancer, 900 patients with predominantly gingival or buccal carcinoma were randomly assigned following surgery to RT alone, accelerated RT, or concurrent chemoradiation with cisplatin 30 mg/m² weekly. For the entire study population, there was no difference in locoregional tumor control, DFS, or overall survival. However, in a post hoc analysis, there was significant improvement in all three outcomes, particularly overall survival, for high-risk patients (T3-T4, N2/N3, ENE) and for patients treated with concurrent cisplatin 30 mg/m² weekly. ³²

Concurrent chemoradiation using carboplatin has also been studied. Carboplatin is more acutely myelosuppressive than cisplatin but results in less nephrotoxicity, neurotoxicity, and nausea and vomiting. Whether carboplatin is as effective as cisplatin as a radiation sensitizer is unclear, as head-to-head randomized trials have not been conducted. At least one trial suggested that every 3 week carboplatin is not as effective as high-dose cisplatin. ³³ However, for patients with renal disease, or limited performance status, or for those unable to tolerate large fluid volumes, weekly carboplatin (AUC 1.5 to 2) remains an alternative to cisplatin. The combination of carboplatin plus fluorouracil may also represent another option when cisplatin is not feasible, although with potential greater toxicity. ³⁴

Overall, it is known that platinum-based therapy with radiation results in increased acute toxicity with an exacerbation of mucositis, increased nausea, and myelosuppression. In addition, cisplatin often results in increased risk of ototoxicity, renal insufficiency, and electrolyte abnormalities. ³⁵

36.3.2 Cetuximab

The EGFR is highly overexpressed in HNSCC and is associated with a poor prognosis. EGFR activation facilitates tumor growth by promoting angiogenesis and proliferation and metastasis by increasing motility and adhesion of tumor cells.

Cetuximab is a chimeric monoclonal antibody (65% human and 35% murine) constructed on an immunoglobulin (Ig) G1 framework, which targets an extracellular epitope in the EGFR ligand-binding domain. ³⁶ Cetuximab blockade of the EGFR results in inhibition of tumor growth, metastasis, DNA damage repair, and angiogenesis. ^{37 , 38}

Several mechanisms that contribute to the antitumor activity of cetuximab have been identified. One is interference by cetuximab with the binding of natural ligands to the receptor itself, thereby disrupting EGFR signaling pathways. ³⁹ Another mechanism involves depletion of the targeted receptors from the cell surface via induction of receptor endocytosis. ⁴⁰ Finally, cetuximab’s construction on an IgG1 framework potentially allows this agent to mediate antibody-dependent cell-mediated cytotoxicity (ADCC) via recruitment of natural killer cells and macrophages. ^{41 , 42}

Preclinical studies have shown that EGFR inhibition sensitizes HNSCC to the effects of ionizing radiation. ^{43 – 46} In mouse A431 xenografts of HNSCC, the combination of cetuximab and cisplatin led to greater growth inhibition than that achieved with either agent alone. ⁴⁷

For the definitive treatment of locally advance disease, it was evaluated in a large international trial of 424 patients with locally advanced HNSCC randomly assigned to RT with or without concurrent weekly cetuximab. ⁴⁶ Median follow-up was 54 months. Overall survival was significantly better for the cetuximab-treated group compared to RT alone (3-year survival 55 vs. 45%, HR = 0.73), as was LRC (50 vs. 41%). It is important to note that this trial compared concurrent cetuximab with RT alone, which is no longer considered a standard approach for locally advanced disease.

For some time, it was not known whether cetuximab and radiation was equivalent to cisplatin and radiation. A retrospective analysis from MSKCC compared the outcome of 125 patients who received cisplatin (100 mg/m² every 3 weeks) with radiation to 50 patients who received cetuximab (400 mg/m² loading dose and 250 mg/m² weekly) with radiation. ⁴⁸ Recognizing that these were two different patient populations, a multivariate analysis to address prognostic imbalances was performed. Despite this adjustment, results for local failure, DFS, and overall survival all favored the cisplatin arm (p < 0.0001 for LF and DFS, p = 0.0017 for OS). A randomized phase II trial that compared concurrent cetuximab plus RT with weekly cisplatin plus RT suggested that cetuximab was associated with more acute toxicity (mucositis, dermatitis, weight loss, and requirements for enteral feeding), more radiation treatment interruptions, and a trend toward worse survival outcomes. ⁴⁹ However, the trial was terminated early and was underpowered to draw definitive conclusions. Given these results though, it is not possible to make assumptions about its role in the adjuvant setting.

RTOG 1016, a large phase III effort which randomized HPV + oropharyngeal cancer patients to cetuximab/radiation versus bolus cisplatin/radiation recently confirmed the survival superiority of the platinum-based regimen, calling into question the future role of cetuximab. ⁵⁰ Additional ongoing studies with cetuximab are discussed below.