OROPHARYNGEAL SQUAMOUS CELL CARCINOMA TREATMENT: CURRENT STANDARDS AND FUTURE DIRECTIONS

Abstract

Purpose of review

To discuss the changing landscape and significant developments in diagnosis and management of oropharyngeal squamous cell carcinoma (OPSCC).

Recent findings

High-risk human papilloma viruses have been recognized as important causative factors for oropharyngeal cancer. The diagnosis is established with type specific and broad-spectrum in-situ hybridization (ISH) probes and/or p 16 immunohistochemistry (IHC) assays on fresh frozen paraffin embedded (FFPE) tissue blocks. HPV associated tumors have superior response and outcomes compared to HPV unrelated tumors. Retrospective studies have been able to stratify OPSCC based on HPV status, tumor stage, nodal stage and smoking history into risk groups with differing risks of death or distant disease. Selected patients, non-smokers with less advanced nodal stage, may be over-treated with current treatment paradigms, and de-intensification of curative therapy is a current research focus for these patients. Smokers, patients with advanced nodal or tumor stage, and HPV-unrelated cancers have a less favorable prognosis and the search for novel targets in particularly important for these patients.

Summary

This review will highlight the current standards and the future direction of novel therapies in both HPV-associated and HPV-unrelated cancers.

Introduction

Worldwide, the incidence of pharyngeal carcinoma is approximately 136 000, with OPSCC on the rise in developed countries.^{(1, 2)} The proportion of all head and neck squamous cell cancers (HNSCC) arising in the oropharynx has risen from 20% in the 1980’s to 70% in the United States at present.^{(3, 4)} This increase has been linked to a rise in HPV-associated cancers.^{(5, 6)} Epidemiologic studies indicate oral HPV infection is predominantly sexually acquired. ⁽⁷⁾ A generational change in sexual behavior may have contributed to the surge in HPV-associated OPSCC.^{(8, 9)}

The scope of this review is to highlight the standards of care for managing oropharyngeal cancer, and how these are forecast to change for HPV-associated and non-associated OPSCC, given a current generation of studies exploring treatment deintensification for favorable prognosis cancers, and testing novel targeted therapies in intermediate and unfavorable prognosis cancers.

The distinct clinical phenotypes, molecular biology and testing of oropharyngeal cancer

Patients with HPV-associated cancer have a younger median age, and lower lifetime exposure to tobacco and alcohol, than patients with HPV-unrelated cancers. [Table 1] ^{(3, 9, 10)} In contrast HPV-negative OPSCC is more common in men in the seventh decade, heavy smokers and in those with a history of alcohol dependence.^{(7, 9, 10)} The two phenotypes are very distinct and Table 1 outlines the differences.

Table 1.

The two distinct subtypes of OPSCC^9,10

	HPV associated OPSCC	HPV negative OPSCC
EPIDEMIOLOGY/RISK FACTORS
1. Race 2. Age 3. Sex 4. Socioeconomic status 5. Smoking /Alcohol History 6. Marijuana use 7. Early sexual debut 8. Multiple lifetime sexual partners	White >Black Between 4th to 6th decade M:F 8:1 Middle to Higher Never or minimal exposure Strong association Strong association Strong association	White > Black Usually 7th decade M: F 3:1 Lower to middle Significant exposure Not known Not known Not known
CLINICAL FEATURES
9. Tumor (T) stage 10. Nodal (N) stage	Early tumor stage More advanced nodal stage	More advanced tumor stage Early nodal stage
OUTCOMES IN STAGE III /IV a,b
11. Distant metastasis risk 12. Second primary (SP) risk 13. Overall response to treatment 14. 2 year OS	Distant control rate : 70–90% Rate of SP :11% >80% respond 95%[ 95% CI, 87–100]	Distant control rate : 70–90% Rate of SP: 4.6% >50% respond 62% [ 95% CI, 49–74]

Illustrates the two distinct subtypes that differ by epidemiology, risk factors, clinical features and outcomes. Abbreviations: M: F Male : Female, DC: Distant Control, SP: Second Primary, OS: Overall Survival

The molecular differences between the HPV-associated and HPV-negative OPSCC suggest distinct pathogenesis.^{(9, 11)} Following viral integration, HPV E6/E7 oncoproteins promote cell cycle progression in keratinocytes by inactivating two important tumor suppressor genes, the tumor protein p53 and Rb. Rapid degradation of p53 results in low levels of intact wild type p53, while Rb inactivation induces the cyclin dependent kinase inhibitor p16.^{(9) (11)} The inactivation of p53 can foster activation of the Wnt signaling pathway⁽¹²⁾ and specific mutations in PIK3Ca have been described in HPV-associated OPSCC.⁽¹³⁾ In contrast, HPV-negative OPSCC are characterized by high mutational burden, with p53 mutation and activation of the PI3K pathway via mutation or PTEN loss the most common abnormalities.^(14–16) The enhanced sensitivity to radiation and chemotherapy of HPV associated cancers is not fully explained.^{(17, 18)} Gene expression differences between these two cancers include frequent p53 mutation in HPV-negative cancers, differential expression of other genes associated with treatment resistance such as TYMS, STMN1, CCND1, and RBBP4; and enhanced immune clearance of virally infected cells following radiation injury.^(19)((20)

HPV DNA is detected by several methods, including type-specific and broad spectrum in situ hybridization (ISH), PCR-based amplification assays, real time PCR to quantify viral load, and IHC for p16, upregulated following oncoprotein E7-induced Rb degradation.^{(21, 22)} p16 IHC is a reliable surrogate marker (sensitivity, 96.8%; specificity, 83.8%) comparable to ISH (sensitivity, 88.0%; specificity, 94.7%)^{(21–23).(24)} The discordance between p16 IHC and HPV ISH for HPV16, the type of HPV most commonly associated with OPSCC, is about 25%. p16-positive /HPV 16 negative cases may be associated with a different HPV genotype. A liquid phase assay used in cervical cancer is being tested with HNSCC.^{(21, 25)}

Risk stratification and identification of favorable oropharyngeal cancer

Prospective and retrospective studies have helped us understand the impact of smoking on oropharyngeal cancer prognosis.

Ang et al retrospectively evaluated a subgroup of 433 patients with oropharyngeal cancer treated on RTOG 0129, a randomized trial comparing accelerated fraction chemoradiation (CRT) to standard fraction RT.⁽²⁶⁾ Tissue was available for HPV testing in 63.8%. The three year overall survival and progression-free survival were significantly better in the HPV-positive than the HPV-negative patients. (HR 0.42 (95% CI 0.27-0.66) p<0.001). A recursive partition analysis (RPA) explored smoking history of ≤ 10 years vs > 10 years, nodal stage and tumor stage, yielding 3 putative risk categories: favorable risk HPV-associated cancer in non-smokers, intermediate risk consisting of HPV-associated cancers with tobacco exposure and high nodal stage, or HPV-negative cancers with intermediate tumor stage, and high risk consisting of HPV-negative cancers with high tumor stage. Table 2 illustrates the differences in outcomes in the various risk for death groups.

Table 2.

Illustrates the OPSCC risk groups from RPA analysis of two retrospective studies and the outcomes.^{(26, 27)}

STUDY	RISK GROUP	VARIABLES	n	3yr OUTCOME (95% CI)
Ang et al. RTOG 0129 (n= 266)25	Low risk for death (n=114)	HPV +, <10py HPV +, >10 py N0-N2a	83 26	OS: 93% (88.3to 97.7)
	Intermediate risk for death (n=79)	HPV +, >10py, N2b-N3 HPV -, <10py, T2-T3	64 15	OS: 70.8% (60.7 to 80.8)
	High risk for death (n= 23)	HPV – >10py, all T4	23	OS 46.2% (34.7 to 57.7)
O’Sullivan et al (n= 505)26	HPV positive low risk for DM	T1–3, N0-N2c	286	DC 93%(89 to 95) LRC 95% (91 to 97)
	HPV positive high risk for DM	T4 N3	63 33	DC 78% (65 to 84) LRC 82% (72 to 89)
	HPV negative low risk	T1-T2, N0-N2c	56	DC 93%(79 to 98) LRC 76% (62 t0 86)
	HPV negative high risk	T3–4, N3	67	DC 72% (56 to 82) LRC 62% (46 to 74)

Abbreviations. RPA: Recursive Partition Analysis, py: pack year, n: Number, HPV : Human Papilloma Virus, OS: Overall Survival, DC: Distant Control, LRC: Locoregional control, T: Tumor, N: Nodal

O’Sullivan et al conducted a similar RPA in a retrospective study with over 800 oropharyngeal cancer patients treated with radiation or concurrent CRT at a single institution between 2001 and 2009.⁽²⁷⁾ HPV status was determined in 56%. HPV-association was demonstrated in 382 patients. RPA stratified patients on HPV status, T stage and nodal stage (Table 2). Smoking status was not identified as a significant risk factor for distant metastases. Conversely, the risk of developing a second primary was significantly higher in patients with > 10 pack year smoking history (11% vs 4%, p=0.009). HPV-associated disease manifested a more indolent natural history, with development of distant metastases occurring until 5 years of follow up, in contrast to most distant failures occurring in the first 2 years in HPV-negative cases.

General principles in the treatment of oropharyngeal cancer

Upfront surgery or radiation alone, for early stage (T1-T2 N0) oropharyngeal cancer are equally effective in loco-regional control and overall survival.⁽²⁸⁾ For patients with early nodal involvement (T1N1, T2N1), resection of the primary with neck dissection or adjuvant radiation is effective. Radiation Therapy Oncology Group (RTOG) study R0920 [NCT00956007] examines whether the addition of cetuximab to radiation in intermediate risk patients without classic indications for chemoradiation improves outcome in the adjuvant setting.

For patients with more advanced T stage and nodal stage (Stage III, IV a and IV b), concurrent high dose cisplatin with radiation has been considered optimal treatment since the publication of the Intergroup trial.⁽²⁹⁾ These findings are also supported by metaanalysis.⁽³⁰⁾ Chemoradiation emerged as the preferred treatment because of the morbidity of conventional surgical approaches to oropharyngeal cancers, which required morbid exposures and generally involved total glossectomy. The advent of transoral surgical approaches for T1 and T2 oropharyngeal cancers has reopened the question of when resection belongs in standard management. Sequential therapy with multiagent induction chemotherapy (IC) followed by standard CRT has been explored as an approach to improve outcome, but randomized trials have failed to demonstrate a benefit in unselected patients.^{(31, 32)}

While organ preservation with concurrent high dose cisplatin with 70Gy radiation is standard of care for locally advanced OPSCC, these patients are at risk for developing long-term toxicities secondary to damage to pharyngeal constrictor muscles, salivary glands, skin and thyroid gland. The risk of dysphagia increases above 55Gy radiation dose to superior and middle pharyngeal constrictors; ⁽³³⁾ feeding tube dependence increases when more than 30% of the pharyngeal constrictors receive 70Gy;⁽³⁴⁾ stricture and aspiration are more common if 50% the pharyngeal constrictors receive 70Gy ⁽³³⁾; Additionally, xerostomia,⁽³⁵⁾ radiation dermatitis,⁽³⁶⁾ chronic cosmetic changes⁽³⁶⁾ and hypothyroidism⁽³⁷⁾ are also associated with higher radiation dose and volume. Long-term follow-up of patients treated curatively for larynx cancer with high dose radiation and high dose cisplatin also raises the possibility that CRT increases non-cancer mortality.⁽³⁸⁾

Future directions in the treatment of oropharyngeal cancer

The HPV-associated oropharyngeal cancer patients are relatively younger individuals, with a lower burden of co-morbidity and excellent prognosis. E2399 treated patients with oropharynx and larynx cancer with induction chemotherapy followed by 70Gy with concurrent paclitaxel, an early attempt at reducing treatment morbidity in unselected patients. ⁽¹⁸⁾ Prospective subset analysis demonstrated that for HPV-positive compared to HPV-negative patients, objective response (81.6 % vs 55.2%, p=0.01) and 2-year progression-free survival were markedly superior (84.2 % vs 56.9%, p=0.06)

Tissue samples from oropharyngeal cancer patients enrolled on Phase III trials RTOG0129 (discussed above),⁽²⁶⁾ TROG 02.02,⁽³⁹⁾ TAX 324,⁽⁴⁰⁾ DAHANCA ⁽¹⁷⁾ all reported that for 3 and/or 5 year survival, HPV positive/p16 positive patients had a significant advantage. (Table 3).

Table 3.

Selected studies with HPV associated oropharyngeal cancer, completed and in progress.

Study	Number	Phase	Treatment	Follow-up (years)	PFS HR (95% CI)	OS HR (95% CI)
ECOG 239917	96	II	IC + CRT with paclitaxel/70Gy	2	0.27 (0.10 to 0.75)	0.36 (0.15–0.85)
RTOG 012925	323	III	Accelerated RT/Cisplatin vs Standard RT/Cisplatin	5	0.49 (0.33 to 0.74)	0.42 (0.27–0.66)
TROG 02.0238	185	III	Cisplatin RT+/- Tirapazamine	2	0.39(0.20–0.74)	0.36 (0.17–0.74)
DHANCA 6 and 716	794	III	Accelerated RT vs Conventional RT	5	DSS 0.47 (0.33–0.67)	0.54 (0.42-0.68)
ECOG 1308	90	II	IC f/b 54 or 69 Gy RT/Cetuximab	1	N/A	N/A
ECOG 3311	377	II	Surgery f/b risk adapted adjuvant therapy	Active	N/A	N/A
RTOG 1016	1000	III	Cisplatin/RT vs Cetuximab/RT	Active	N/A	N/A

Abbreviations: PFS: Progression free Survival, DSS : Disease Free Survival, OS : Overall Survival, HR : hazard ratio, CI :confidence interval

I: HPV associated disease: De-escalation clinical trials in progress

Given the difference in etiology, molecular profile, natural history and treatment responsiveness, investigators have concluded that HPV-associated and HPV-unrelated oropharyngeal cancer are two different diseases, and that different research questions arise in the two diseases. For HPV-positive cancer, it will be important to develop reliable tests – whether based on clinical criteria or on validated biomarker panels – to select patients with a very high chance of cure, and to find treatments for these patients which are less arduous, and which are associated with the least possible risk of long term toxicity and of non-cancer mortality. For HPV-associated cancer with a greater risk of recurrence, defining the causes of failure and adapting treatments to address these causes will be important. For the HPV-negative locally advanced patient, who currently appears to have only a 35% chance of cure, identification of novel targets, and incorporation of new agents, will be critical to improved outcomes.

Several strategies have been explored.

1. De-escalating treatment with upfront induction chemotherapy followed by response adapted radiation (low vs standard) with an alternative radiosensitizer

E1308[NCT01084083], a prospective Phase II trial designed specifically for HPV16 ISH or p16 positive, stage III, IVa oropharyngeal cancer patients, completed accrual in 2011. (ASCO 2013, Abstract 6005) Treatment included IC with three cycles of cetuximab/paclitaxel/cisplatin to identify the most treatment responsive patients, who then received 54 rather than 69 Gy radiation with concurrent cetuximab. With a median follow up of 16.2 months, the one-year PFS in low dose arm in patients with ≤10 pack year smoking history patients was 97% (95% CI: 0.83 to 0.99). Follow on studies planned within ECOG include a trial to examine whether the dose to next echelon nodes can be reduced, what the least toxic induction regimen to identify the 70% of patients who are eligible for radiation de-escalation is, and whether cisplatin or cetuximab is the preferable radiosensitizer. These developmental studies are anticipated to lead to an optimal low toxicity approach to deintensification which will be appropriate for phase III testing against conventional full dose CRT with high dose cisplatin. RTOG proposes to test treatment deintensification from 70 to 60Gy with no radiation sensitizer in a comparable population.

2. Deescalating treatment by optimizing the type of radiosensitizer with standard dose radiation

RTOG 1016 [NCT01302834], a Phase III trial comparing weekly cetuximab with accelerated RT 70Gy/6 weeks to cisplatin 100mg/m2 d1 and 22 with accelerated RT 70Gy/6 weeks in patients with p16 positive, T1-2, N2a-N3 and T3/4 any N, oropharyngeal cancer reached its initial accrual goal in mid-2013, but proved to have a lower event rate than projected. It has recently reopened with an increased sample size of 1000. This study is the first trial to compare cisplatin with cetuximab as a radiosensitizer in a homogenous population.

3. Deescalating treatment using upfront surgery with minimally invasive techniques followed by reduced dose radiation in the adjuvant setting

Previously surgery for oropharyngeal cancer required transcervical neck exposure and resulted in significant morbidity. Minimally invasive techniques for transoral have emerged.⁽²⁸⁾ To date, the experience with these techniques, specifically on tonsillar and base of tongue T1 and T2 OPSCC have reported excellent long term swallowing outcomes. ECOG 3311 [NCT01898494] is a Phase II trial in p16 positive resectable Stage III, IV a and b oropharyngeal cancer. All patients will undergo transoral resection and neck dissection, followed by risk adapted adjuvant therapy. Patients with pathological Stage I and II, are considered low risk for recurrence and will enter the observation arm. Patients with <1mm extracapsular extension (ECE), 2–4 positive lymph nodes or close margins, will be considered intermediate risk. These patients will be randomized to 50Gy vs 60Gy radiation. The high-risk group are those with positive margins, with >1mm ECE, or at least 5 positive lymph nodes. These patients will receive 66 Gy with weekly cisplatin. This study hopes to demonstrate that transoral resection followed by reduced dose adjuvant radiation can maintain a 2 year PFS in the intermediate risk group of 85% or better.

II: HPV Negative Disease: Treatment intensification with novel targeted therapy

As noted above, decades of trials examining longer duration and intensity of therapy,^{(31, 41)} accelerated fractionation in chemoradiation, and the addition of novel agents to CRT, ^{(26, 42)} have failed to advance survival for patients with HPV-negative cancer. Data from a randomized phase II trial suggest the possibility that the dual EGFR/HER2 inhibitor lapatinib added to CRT improves disease control and survival in HPV-negative HNSCC, and has led to the design of an RTOG Foundation study of lapatinib and CRT in p16-negative HNSCC [NCT00490061].⁽⁴³⁾ Another dual EGFR-HER2 targeted agent is the irreversible inhibitor afatinib. This tolerable and orally administered agent is active in HNSCC and opened up the opportunity for long term adjuvant therapy in patients who have no evidence of disease following CRT, but are at high risk of recurrence by virtue of tobacco history or stage and subsite. These patients are eligible for 2:1 randomization between afatinib and placebo on a trial powered for disease-free and overall survival [NCT01427478]. Additional novel targets of interest in this population are Aurora A kinase [NCT01540682]; ⁽⁴⁴⁾ tubulin; ⁽⁴⁵⁾ EGFR ⁽⁴⁶⁾ and the PI3K pathway, now recognized to be the most commonly deranged pathway in these cancers [NCT01111058; NCT01602315; NCT01737450]. [Table 4]

Table 4.

Selected studies in progress with targeted therapies in HNSCC

Study	Number	Phase	Treatment	Targets
NCT00490061	60	II	Lapatinib+RT	EGFR and Her-2/neu expressions
NCT01427478	315	III	Afatinib vs Placebo after CRT	EGFR and Her-2/neu expressions
NCT01540682	9-18	I	MLN8237 + Cetuximab + RT	Aurora A and EGFR expression
NCT01111058	160	II	Everilomus vs Placebo after CRT	PI3K pathway
NCT01602315	186	Ib/II	BYL719 and Cetuximab in metastatic/refractory	PI3K pathway
NCT01737450	70	II	BKM120 in metastatic/refractory	PI3K pathway

Abbreviations: RT: Radiation therapy, CRT: Chemoradiation therapy

An important new direction in anticancer therapy has been the ability to target T cell exhaustion. Both in HPV-associated and HPV-unrelated HNSCC, it seems likely that a subset of PD-1 expressing cancers will benefit from this approach, with phase I trials ongoing and trials through phase III currently under development.

Conclusions

The oropharyngeal cancer landscape is swiftly changing and the number of HPV associated oropharyngeal cancers are expected to surpass the number of cervical cancers by 2020.⁽⁴⁾ Investigators now hope to devise different treatment paradigms for optimal management of OPSCC based on HPV status, tumor stage, nodal stage and smoking history. Important research questions will be the most accurate means of identifying highly curable patients, the least toxic treatment approaches which do not sacrifice cure in favorable risk patients, and identification of new agents and paradigms for patients who currently do not have a high expectation of cure.

KEY POINTS.

1. Oropharyngeal carcinoma is on the rise in developed countries.

2. High risk human papilloma viruses (HPV) have been recognized as important causative factors for oropharyngeal cancer.

3. HPV associated OPSCC has a distinct clinical and molecular biology compared to HPV negative OPSCC.

4. Treatment response and survival outcomes in HPV associated OPSCC are superior compared to HPV negative OPSCC.

5. Clinical trials focus on de-escalation of treatment in select group of HPV associated OPSCC, and intensification of treatment with addition of novel targeted agents to standard treatment in HPV negative OPSCC.