An epidemiological insight into HPV status and prognostic gene mutations in head and neck cancer

A study release from Clarivate:

Summary

Since human papilloma virus (HPV) infection is related to sexual practices and HPV positive head and neck cancers (HNCs), particularly in men, are on the rise, several studies have documented the role played by high-risk HPV infections in these cancers [1]. HPV is a risk factor for HNC; somatic mutations are less frequent in HPV positive HNC; and finally, HPV positive cancers respond better to treatment and have better outcomes.

Associating differentially expressed genes in HNCs with etiology (alcohol/smoking/HPV) identifies prognostic genes. HPV negative HNCs represent a distinct cancer type with an inferior outcome. In Africa, the major populations affected by HNCs are HPV negative, with a poor prognosis. However, a survival advantage (55% decrease in death risk) is noted here, in the presence of NSD1 mutations [2, 3]. The protein lysine methyltransferase NSD1 is a key regulator of the epigenome. In HPV negative tumors, the survival advantage from NSD1 mutations can be partly explained by increased sensitivity to DNA crosslinking agents including carboplatin. Protein methyltransferases such as NSD1 show potential as the next generation therapeutic targets in HNCs.

In this report, Shyama Ghosh (Ph.D.) reviews the prevalence of NSD1 and other differentially expressed genes, as well as ongoing clinical studies targeting specific mutations, in HPV positive versus HPV negative HNC patients in Africa, Europe and the U.S. 

Global burden of head and neck cancers

Head and neck cancers (HNCs) are mostly squamous cell carcinomas and include cancers of the oral cavity, oropharynx, hypopharynx, and larynx. According to the GLOBOCAN project, HNCs are the sixth leading malignancy with an estimated annual incidence of about 633,000 and 355,000 deaths worldwide [4]. When the classification includes esophageal cancers, the category is termed upper aerodigestive tract cancer (UADTC; or HNC + esophageal cancers).

GLOBOCAN 2020 ranks UADTC and HNC incidence by world regions and countries (Table 1.) [4].

Table 1. HNC and upper aerodigestive tract cancer (UADTC; HNC plus esophageal cancer) ranking for countries and regions based on the age-standardized incidence rates (ASIR). ASIR is presented per 100,000 per year [4].

African regions have high incidence rankings for both. UADTC ranks first in east Africa, while HNC ranks second for Sudan. UADTC cases are also numerous in Southern Africa.  Overall, in Africa, the annual number of new cancers of the oral cavity, larynx, and oropharynx number at 13,613, 10,058 and 2,514 respectively [5].  For the year 2050, the projected number of new oral and lip cancer cases in Africa is forecast as follows: sub-Saharan Africa, 57,327; east Africa, 13174; west Africa, 11968 and south Africa, 3121 [4, 6].

In contrast, low rankings are noted for the U.S. and North America – sixth and fifth respectively [7]. Over the years Europe and the U.S. have seen a decline in this tumor prevalence thanks to anti-smoking campaigns and health policies. In contrast, a gradual rise in HNCs caused specifically by Human Papillomavirus (HPV) infection is noted in these regions, especially impacting the oropharyngeal cancers (OPSCCs).

Globally, the burden of HNC and UADTC is high, with less developed regions ranked higher than the more developed ones (age-standardized incidence rate [ASIR], HNC, 8.4 vs 3.6 [per 100,000 per year]). Rates of esophageal cancer are also higher in the less developed world than the developed counterparts (ASIR, esophageal cancer, 7.2 vs 1.5). Treatment relies on combinations of surgery, radio- and chemotherapy and the overall 5-year survival rate is about 60%.

Differentially expressed genes in head and neck cancers

To unravel the molecular biology of these cancers, the Cancer Genome Atlas (TCGA) and COSMIC (Catalog of somatic mutations in cancer) provide the largest publicly available HNC dataset for dominant genetic mutations in the U.S. [8, 9]. The TCGA presents a thorough genomic characterization of 279 HNCs, identifying molecular changes in a subset of oral cavity (62%), larynx (26%) and oropharynx tumors (12%) [8, 10]. The HNC genome bears an overall mean of 141 copy number alterations (CNAs; amplifications or deletions) and 62 chromosomal fusions per tumor.

Table 2. displays the significantly mutated genes identified in the 279 HNCs. The Nature study and a follow-up analysis of the TCGA HNC data confirm the high number of somatic mutations in TP53 followed by FAT1 [8, 10, 11].  Co-occurrence pattern of CDKN2A, TP53 and FAT1 mutations and a mutually exclusive expression pattern for TP53 and PIK3CA mutation are recorded.

Table 2. Differentially-expressed genes (DEGs) identified in HNCs using the MutSigCV algorithm (q < 0.1) ordered by q value (TCGA and COSMIC [upper aerodigestive tract tissue]) databases [8-10].

Genetic marker-driven clinical trials

The following clinical studies are reported:

• Pan FGFR kinase inhibitor BGJ398 (Infigratinib) to treat patients with FGFR1-3 translocated, mutated, or amplified recurrent head and neck cancer (NCT02706691).
• Phase II study of Tipifarnib in patients with HRAS mutations (NCT02383927).
• Copanlisib in association with Cetuximab in recurrence and/or metastasis harboring a PI3KCA mutation/amplification and/or a PTEN loss (NCT02822482).
• SF1126 in recurrent or progressive cancer with mutations in PIK3CA gene and/or PI-3 kinase pathway genes (NCT02644122).
• Korean Cancer Study Group: Translational bIomarker Driven UMbrella Project for Head and Neck (TRIUMPH), Esophageal Squamous Cell Carcinoma- Part 1 (HNSCC) (NCT03292250).

• An umbrella biomarker study (phase II) has been designed for recurrent and/or metastatic HNSCC patients (EORTC-1559-HNCG, NCT03088059). Six biomarker driven patient cohorts and their planned medication are reported as follows: p16 negative and EGFR amplification/mutation or PTEN high or HER2 amplification/mutation, Afatinib; p16 negative and cetuximab naïve, Afatinib; p16 negative and CCND1 amplification, Palbociclib; p16 negative and platinum sensitive, Niraparib; p16 positive oropharyngeal cancer, Niraparib; FGFR1/2/3 mRNA overexpression, Rogaratinib.

The first three cohorts will also have access to standard of care (Methotrexate, Paclitaxel, Docetaxel, Carboplatin, 5-Fluorouracil, Bleomycine, Gemcitabine, Mitomycine or Best supportive care). An immunotherapy cohort will receive Monalizumab while another will be given Monalizumab plus Durvalumab. Patients included in the Afatinib arms should not have activating mutation in RAS. Patients included in the Rogaratinib arm should not have activating mutation in RAS or PIK3CA.

The EORTC1559 umbrella trial is open for inclusion since December 2017. The EORTC-1559-HNCG trial is the first European international umbrella trial assessing a personalized treatment strategy for patients with recurrent/metastatic HNSCC. The protocol was submitted in four different countries (Belgium, France, Italy, and UK) and will be submitted in Germany to both competent authorities and applicable ethics committees.

Etiology of head and neck cancers

The etiology of HNC varies by geographic region. While smoking and/or alcohol abuse are known to cause HNCs in Africa, human papillomavirus (HPV) infection is largely responsible for HNCs in the U.S. and Europe. HPV is a family of closely related non-enveloped double-stranded DNA viruses. It is the most widespread sexually transmitted infection, with high- and low-risk subtypes that are based on oncogenic potential. Among the high-risk subtypes, HPV16 accounts for over 82% of all HPV-positive cases, 50% of all cervical cancer cases and a significant proportion of HNCs.

U.S. population data from the CDC’s National Program of Cancer Registries (NPCR) and the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) Program (2012-2016) have shown that of the 19,000 oropharyngeal cancers reported per year (females 3,460; males 15,540), up to 70% are caused by HPV (females 63%; males 72%). Another study of 1,374 pharyngeal cancers, 1,264 oral cavity cancers, and 1,042 laryngeal cancers from 29 countries, confirmed the presence of HPV DNA in 22.4% of oropharyngeal cancers, 4.4% in oral cavity cancers, and 3.5% in laryngeal cancers.

Meta-analysis of 148 studies involving 12,163 HNC cases showed the existence of HPV DNA in 31.5% of tumors with greater prevalence in the OPSCCs (45.8%), followed by oral squamous cell carcinomas (24.2%) and laryngeal squamous cell carcinomas (22.1%) [5].

In the TAX 324 trial, 59/68 (87%) subjects with HPV(+)HNC had oropharyngeal squamous cell carcinoma (OPSCC) [12, 17]. 

Prognosis in HPV caused head and neck cancer

HPV+ oropharyngeal squamous cell carcinoma (HPV[+]OPSCC) is associated with oral sexual behaviors and is more prevalent among younger individuals. Patients with HPV associated HNC, particularly those with OPSCC, show better treatment response, higher survival rates, and lower risk of recurrence as compared to patients with HPV(−)HNC (cancer caused by alcohol or tobacco use).

Following standard chemotherapy, clinical studies such as the TAX 324 trial observed better overall survival of HPV(+)HNC patients as compared to stage-matched HPV(−)HNC patients, with 5-year survival rates of 60%-90% versus 20%-25%. Median overall survival for all HPV(−)HNC patients and HPV(−)OPSCC patients were 26.6 and 19.7 months respectively [12, 17].

“HPV status is now declared to be an independent predictor of overall survival in patients with recurrent or metastatic HNC. This calls for dissociating the genetic marker profile according to HPV status in HNCs”.

Genetic marker distribution in HPV(+) versus HPV(−)HNC

The Cancer Genome Atlas (TCGA) provides US data on genetic mutations in HPV(+)HNC that together with HPVp16 or the E6/E7 proteins indicate good prognosis (Table 3.) [8]. Differentially expressed genes are identified by stratifying according to HPV status. The multi-platform profiling study includes 36 HPV(+) and 243 HPV(−) tumors.

Table 3. Key genes altered and oncogenic events listed by HPV status in HNCs [10]. Genes marked in purple display aberrant activity.

HPV(+) tumors exhibit infrequent mutations in TP53, an intact CDKN2A oncogene and activating mutations of the oncogene PIK3CA. Such carcinomas also demonstrate loss of the TNF receptor-associated factor 3 (TRAF3) and amplification of the cell cycle gene E2F1. These alterations promote the aberrant activation of NF-kappa signaling that may play a role when targeted therapies are developed for these tumors.

In contrast, HPV(−)HNCs express homozygous deletion of the TP53 gene at a high rate (84% mutation rate) and the CDKN2A gene is deleted. Recurrent focal amplifications in receptor tyrosine kinases (for example, EGFR, ERBB2 and FGFR1) also predominate in HPV(−)tumors, while gain-of-function mutations in the NFE2L2 (NRF2) pathway (key transcription factor regulator of oxidative stress) are highly prevalent in HPV(−)HNCs but are rare in HPV(+) tumors.

Genetic markers indicating good prognosis

In HPV(−)HNC, high NFE2L2 (NRF2) activity may regulate the expression of antioxidant proteins. However, the loss of p53 function and deletions in the tumor suppressor genes (TP53, CDKN2A) in these tumor correlate with chemoresistance, radio resistance and therefore a poor prognosis.

In HPV(+)HNC, combination of HPV status (p16) with near normal expression of the receptor tyrosine kinases (EGFR, FGFR1, IGF1R), oncogenes, tumor suppressor genes (TP53, CDKN2A) and mutated activity from FGFR3-TACC3 fusion and PIK3CA amplification are noted. Further detailed study of the above genes can stratify HPV(+)HNCs into good versus poor responders to treatment, thereby guiding de-escalation treatment approaches in such patients (as studied in the De-ESCALaTE HPV trial) [13].

HPV infection/surrogate markers of the infection are clinically relevant prognostic indicators in OPSCC. It is important though to be certain that the active HPV infection is causing the cancer. Some of the diagnostic methods used for this are the PCR-based detection of E6/7 mRNA, L1 mRNA or HPV16 DNA detection and viral load analysis, HPV16 DNA diagnosis by in situ hybridization, E6/7/L1 serum antibodies and the MassARRAY system of combining PCR with mass spectrometry to determine viral load and gene expression.

In function, HPV E6/E7 proteins sequester the tumor suppressor p53 and the retinoblastoma (Rb) gene product pRb that results in overexpression of the p16INK4A protein via a negative feedback mechanism.

p16INK4A is associated with overall and progression-free survival in OPSCC.

Yet, even in those OPSCCs where HPV infection is confirmed, good prognosis may also depend on HPV copy number, smoking status, p53, and EGFR. 

EGFR amplification and poor prognosis in HNC

EGFR (ErbB1/HER1) is a member of the ErbB tyrosine kinase family, which also includes HER2 (ErbB2),

HER3 (ErbB3) and HER4 (ErbB4) [14]. EGFR is over expressed in many human malignancies including up to 90% of head and neck squamous cell carcinomas (HNSCC). Moreover, tumors with a high EGFR gene copy number and/or protein over expression have been strongly associated with worse clinical outcomes.

Cetuximab, manufactured in Germany and the U.S., is a monoclonal chimeric (human/mouse) antibody that targets EGFR with a twofold mechanism of action: direct EGFR binding to stop its activation and induction of antibody-dependent cellular cytotoxicity to clear antibody-coated cells. Approved by the U.S. and European regulatory agencies, cetuximab is used in locally advanced HNSCC.

EGFR alterations (high gene copy numbers, overexpression) are inversely correlated to HPV status in OPSCC. HPV status and EGFR expression are two well established prognostic factors yet while the first is associated with improved outcomes, the latter is associated with poor prognosis.

An inverse correlation between EGFR expression and HPVp16+OPSCC is noted. HPV(+)OPSCCs, defined as HPV DNA or p16+, express significantly less EGFR. A significant difference in EGFR gene copy numbers is observed according to HPV status; an increased EGFR gene copy number is largely restricted to p16−OPSCCs.

The situation in Africa – comparison with US data

The data described till now mainly involves white Caucasian patients from developed countries. Since the etiology of HNC varies by geographic region and HPV(+) cases have been mostly studied in the U.S. the extent of this infection was next studied in the developing world.  In Africa, HNCs are usually caused by smoking or alcohol abuse. Prevalence of HPV(+)HNC is low, with a South Africa study of oropharynx, nasopharynx, hypopharynx, and laryngeal tumors reporting HPV prevalence of only 6.3% [15].

The TAX 324 trial showed that the proportion of HPV(+) tumors is nearly 9-fold higher in white patients (66 of 196, 34%) than in black patients (1 of 28, 4%) [12, 17]. Another sub-Saharan Africa study recorded the presence of HPV(+) DNA in 19.23% of HNC. HPV E6/E7 oncogenic DNA was found in 18% of the HNSCC cases, with HPV16 being the predominant genotype present [5].

Gender disparity

Oropharyngeal cancer (OPSCC) displays a 3:1 ratio for males to females. Studies on Sub-Saharan Africa (1990-2013) included 29 papers on squamous cell carcinomas (SCCs) of the oral cavity/oropharynx, with 7,750 patients [16]. The mean age of patients ranged from 37 years in Kenya to 58 years in Ghana. The youngest mean age in a generalized oral cavity/oropharynx SCC population was 46 years from Congo. Male-to-female ratios ranged from 0.5:1 in Congo to 4:1 in South Africa. Overall, there were 2.3 males per female with oral cavity/oropharynx SCC across all studies.

Outcomes by Race

HPV plays an important role in racial disparity among HNC patients. Settle described the incidence of HNSCC to be higher in black patients than in white patients [12]. Data from the U.S. Surveillance, Epidemiology, and End Results database showed that black patients present with more advanced disease and have twice the age-adjusted mortality rate compared with white patients. Even after correcting for stage at diagnosis, black patients had a significantly worse survival than white patients.

Oral cancer for example has a high age standardized mortality rate (ASMR) of 6.8 (per 100,000 persons per year) in black men as compared to white or Hispanic counterparts [7].

HPV and racial disparity

Disease-free survival is significantly higher in white than in black HNC patients who are treated with chemoradiation; the greatest difference occurs in the OPSCC subgroup. HPV, primarily HPV16 infection is associated with a significant percentage of OPSCCs, primarily of the base of the tongue and tonsil and HPV(+) cancers have a significantly better prognosis than do the HPV(−) diseases. HPV thus plays an important role in racial disparity within HNC.

Associating race and HPV status with response to therapy and survival

The phase 3, multicenter TAX 324 trial of induction chemotherapy followed by concurrent chemoradiation in HNC patients showed that median overall survival (OS) in the retrospective cohort was 52.1 months among white patients compared to only 23.7 months in their black counterparts (Table 4.) [17]. This disparity was entirely due to the subgroup of patients with OPSCC (OS of 69.4 months [white patients] versus 25.2 months [black patients]).

Table 4. TAX324 trial survival outcomes by race, tumor site and HPV status [17].

When assessing by HPV status, HPV positivity was 34% in white patients versus only 4% in black patients (9 folds higher). median OS was significantly worse for black patients (20.9 months) than for white patients (70.6 months) and dramatically improved in HPV(+) versus HPV(−) (26.6 months, 5.1 hazard ratio) OPSCC patients, 49% of whom were HPV16 positive. Survival was similar for black and white HPV(−) patients.

TAX 324 confirms HPV status in black versus white HNSCC patients [17]. These findings have important implications for the etiology, prevention, prognosis, and treatment of HNSCC. Moreover, the presence of a predominant HPV(−) phenotype among black HNSCC patients drives the need to characterize this population for differentially expressed genes.

“The TAX 324 study indicates that a large percentage of black HNC patients are HPV negative: 96% as compared to 66% of white patients”.

Protein methyltransferases in HNSCC

Whole exome sequencing identified numerous gene expression changes in the protein methyltransferases (PMTs) in HNSCC [18]. The nuclear receptor-binding SET domain protein 1 (NSD1) gene, a methyltransferase and chromatin modifier, is among the top 10 most frequently mutated genes in HNSCC, with most mutations predicted to be inactivating [2]. Kaplan-Meier survival analysis showed significantly improved survival in HNC patients with NSD1 gene alterations when compared to patients with wild-type NSD1 tumors [2]. Increased sensitivity to platinum-based chemotherapy agents associated with NSD1 depletion (via ERCC5 inactivation) may contribute to improved survival [19]. In the quest for prognostic mutations in HPV(−)HNC cases, NSD1 is a potential target. More than 25% of the tumors of larynx and oral tongue are HPV(−), thus there is a high probability for enrichment of NSD1 mutations here [2].

Discussion

HNSCC is the 6^th most common malignancy worldwide with roughly 500,000 new cases leading to 300,000 deaths each year [18]. It comprises cancers of the epithelium of the oral cavity, tonsils, pharynx (including the nasopharynx, oropharynx, and hypopharynx), larynx, epiglottis, the paranasal sinuses, and the nasal cavity. HNSCC is pathogenetically classified as HPV(+) and HPV(−) disease.

Standard surgery and/or chemoradiotherapy treatment options are associated with 5-year recurrence rates of about 50% in HPV(−)HNSCC. Survival numbers are low and second line standard of care treatment with immunotherapy is frequently followed. Yet this line of treatment by checkpoint inhibition benefits a small fraction of patients.

The Cancer Genome Atlas (TCGA) demonstrates a molecular analysis of HPV(+) and HPV(−) tumors, identifying recurrent mutations in 11 genes such as TP53 (72%), FAT1 (23%), CDKN2A (22%), NOTCH1 (19%) and NSD1 (10%) [2, 7].

Next generation sequencing (NGS) technologies have identified potential targets that are genomic alterations in HNSCC [20]. Targetable genomic alterations in HPV(−)HNSCC include those related to kinase growth factor family receptors such as EGFR (15%), FGFR1-3 (2%-10%), HER2 (5%), phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha (PIK3CA) (34%) and HRAS (5%). HPV(−) tumors can also bear potential actionable cell cycle genomic alterations: TP53 mutation (84%), cyclin D1 (CCND1) amplification (31%), and CDKN2A inactivation (58%). In contrast, HPV(+) tumors display inactivation of p53 and Rb by the oncoproteins E6 and E7. In these tumors, PIK3CA amplifications/mutations are found in 56% of cases while other genomic alterations are rare.

Since HPV(+) tumors have a more favorable prognosis than HPV(−) cases (74% vs. 30%, 5-year overall survival rate in stage IV disease), investigators aim at de-intensifying therapy in HPV(+) while exploring novel therapeutic approaches for HPV(−) tumors [3]. For both disease types, current standard of care for localized HNSCC involve surgery, radiation, and concomitant cisplatin chemotherapy. Other approaches include combination chemotherapy and EGFR inhibition, with mixed results. Platinum-based chemotherapy together with cetuximab is effective as first-line treatment, with Nivolumab increasing overall survival after platinum therapy.  Pembrolizumab is also approved for the same indication by the U.S. Food and Drug Administration.

Despite multimodal treatment regimens of surgery combined with chemoradiation, less than 60% of HNSCC patients are disease-free at 3 years while those with recurrent or metastatic disease not undergoing radiotherapy or surgery have a median survival of 10-12 months [20]. For immunotherapy, programmed cell death protein 1 (PD-1)/PD-L1 blockers have activity in HNSCC but the 2-year overall survival rate is still low at 16.9%. No standard of care exists for patients who progress after platinum-therapy and anti PD-1 compounds.

HPV(−)HNSCC represent distinct cancer type with worse expected outcomes. However, survival advantage (55% decrease in death risk) is noted in the presence of NSD1 mutations [2, 3]. Patients with NSD1 mutations reported a significantly improved outcome (approximately 5-year increase in median overall survival time [8.0 vs. 3.1 years]) [2]. NSD1 seems to play a causal role in these associations, as disrupting NSD1 in vitro leads to CpG hypomethylation and sensitivity to cisplatin, with up to 50% decrease in IC50 values [2].

Protein lysine/arginine methyltransferases may have potential as the next generation therapeutic targets in HNSCCs.

References

1. Mendez. (2012) Biomarkers of HPV Infection in Oropharyngeal Carcinomas: Can We Find Simplicity in the Puzzle of Complexity?
2. GDC Data Portal (NIH) NSD1 mutations (curated on February 20, 2021).
3. Bui N et al. (2018) Disruption of NSD1 in head and neck cancer promotes favorable chemotherapeutic responses linked to hypomethylation.
4. GLOBOCAN 2020. Referred to February 20, 2021.
5. Aboagye E et al. (2019) Human Papillomavirus Detection in Head and Neck Squamous Cell Carcinomas at a Tertiary Hospital in Sub-Saharan Africa.
6. Hille and Johnson. (2017) The burden of oral cancer in sub-Saharan Africa: An estimate as presented to the Global Oral Cancer Forum, March 2016.
7. Adeola H et al. (2018) The burden of head and neck cancer in Africa: the status quo and research prospects.
8. The Cancer Genome Atlas Program (TCGA). Referred to February 20, 2021.
9. Catalog of Somatic Mutations in Cancer (COSMIC). Referred to February 20, 2021.
10. The Cancer Genome Atlas Network. (2015) Comprehensive genomic characterization of head and neck squamous cell carcinomas.
11. Sayáns M et al. (2019) Comprehensive Genomic Review of TCGA Head and Neck Squamous Cell Carcinomas (HNSCC).
12. Settle K et al. (2009) Racial Survival Disparity in Head and Neck Cancer Results from Low Prevalence of Human Papillomavirus Infection in Black Oropharyngeal Cancer Patients.
13. Ramesh P et al. (2020) NRF2, p53, and p16: Predictive Biomarkers to Stratify Human Papillomavirus Associated Head and Neck Cancer Patients for De-Escalation of Cancer Therapy.
14. Mirghani H et al. (2014) Oropharyngeal cancers: Relationship between epidermal growth factor receptor alterations and human papillomavirus status.
15. Sekee T et al. (2018) Human papillomavirus in head and neck squamous cell carcinomas in a South African cohort.
16. Faggons C et al. (2015) Review: Head and neck squamous cell carcinoma in sub-Saharan Africa.
17. TAX324 trial
18. Saloura V et al. (2018) The role of protein methyltransferases as potential novel therapeutic targets in squamous cell carcinoma of the head and neck.
19. Pan C et al. (2019) NSD1 mutations by HPV status in head and neck cancer: differences in survival and response to DNA-damaging agents.
20. Galot R et al. (2018) Personalized biomarker-based treatment strategy for patients with squamous cell carcinoma of the head and neck: EORTC position and approach.