Abstract
While P16 immunohistochemistry (IHC) is a well-established surrogate marker of Human Papillomavirus (HPV) in oropharyngeal squamous cell carcinoma (OSCC), Retinoblastoma 1 (RB1) loss may lead to p16 overexpression in the absence of HPV. We determined the proportion of p16-positive/HPV-negative OSCC with RB1 loss and other alterations in RB1/p16 pathway, and tested RB1 IHC as a prognostic biomarker for OSCC, along with the 8th edition of AJCC staging manual. P16 and RB1 IHC and HPV DNA in situ hybridization (ISH) were performed on 257 OSCC. High risk HPV RNA ISH, RB1 fluorescence in situ hybridization (FISH), and next generation sequencing (NGS) were done on p16-positive/HPV DNA ISH-negative OSCC. Disease free survival (DFS) was used as an endpoint. In the entire cohort and in p16-positive (n = 184) and p16-negative (n = 73) subgroups, AJCC 8th edition staging correlated with DFS (p < 0.01). RB1 IHC showed RB1 loss in p16-positive OSCC only (79/184, 43%). RB1 loss by IHC is associated with a better DFS, without providing additional prognostic information for patients with p16-positive OSCC. HPV RNA ISH was positive in 12 of 14 HPV DNA ISH-negative cases. RB1 IHC showed loss in 10 of 15 HPV DNA ISH-negative cases and in 1 of 2 HPV RNA ISH-negative cases. Overall, only one case of p16-positive/HPV RNA ISH-negative OSCC showed RB1 loss by IHC (1/184, 0.5%). Of the 10 p16-positive and HPV DNA ISH-negative cases with RB1 loss by IHC, 2 had RB1 hemizygous deletion and 3 showed Chromosome 13 monosomy by FISH. No RB1 mutations were detected by NGS. Other molecular alterations in p16-positive/HPV DNA ISH-negative cases included TP53 and TERT mutations and DDX3X loss. HPV-independent RB1 inactivation rarely results in false positive p16 IHC. RB1 inactivation by high risk HPV E7 oncoprotein may co-exist with RB1 deletion. RB1 loss is a favorable prognosticator and occurs exclusively in p16-positive OSCC. The 8th edition of the AJCC staging manual satisfactorily predicts DFS of OSCC patients.
Keywords: Oropharyngeal squamous cell carcinoma, Retinoblastoma 1, RB1, Human papillomavirus, P16, AJCC 8th edition staging classification
Introduction
Human papillomavirus (HPV) mediated oropharyngeal squamous cell carcinomas (OSCC) are characterized by a more favorable outcome and are staged following distinct TNM criteria [1–3]. In addition to being a staging and prognostic factor, HPV status is important in the evaluation of cervical lymph node metastases with squamous cell carcinoma of unknown primary (SCCUP), as HPV positivity is commonly indicative of oropharyngeal origin [4].
P16, a cell cycle regulator protein encoded by CDKN2A gene (cyclin-dependent kinase (CDK) inhibitor 2A), is overexpressed in HPV-positive OSCC and p16 immunohistochemistry (IHC) is accepted as a surrogate biomarker for HPV in OSCC. In HPV-positive OSCC, p16 overexpression is a result of tumor suppressor protein Retinoblastoma 1 (RB1) inactivation by high risk HPV E7 oncoprotein [5–8]. However, p16 overexpression may also be HPV-independent, as a result of p16/CDKN2A, RB1, TP53, or cyclin-dependent kinase 6 (CDK6) alterations [9, 10]. It appears that up to 7.7% of p16-positive OSCC are HPV negative and show loss of RB1 [5, 11]. The prevalence and significance of such genetic alterations in OSCC and their direct correlation with p16 and RB1 IHC are unclear, but it has been reported in HPV-negative extra-oropharyngeal tumors (e.g., p16 IHC positive in neuroendocrine carcinomas and cutaneous squamous cell carcinomas due to RB1 inactivation) [12, 13]. In the scenario of metastatic SCCUP and when the possibility of cutaneous primary is considered, 25% of cutaneous squamous cell carcinomas may show positive p16 IHC secondary to HPV-independent RB1 loss [13]. For these reasons, for SCCUP, direct methods of HPV detection, usually by DNA or RNA in situ hybridization (ISH), are recommended [3, 4, 9].
The role of RB1 alterations as an additional prognostic marker in OSCC is unclear. Some studies showed that RB1 loss is associated with a better prognosis, p16 overexpression, and presence of HPV [5, 14, 15]. On the other hand, Beck et al. described an aggressive metastatic p16-positive OSCC with ambiguous HPV status and RB1 mutation [16]. In the same study, The Cancer Genome Atlas (TCGA) and Foundation One database analyses revealed that patients with OSCC characterized by high p16 and low RB1 had a decreased survival [16].
Therefore, we decided to determine how frequently RB1, p16/CDKN2A, TP53, and CDK6 alterations alone, in the absence of HPV, may lead to p16 overexpression (i.e., false positive p16 as surrogate HPV biomarker). To this end, we have explored p16 and RB1 status first by IHC in a cohort of 257 patients with OSCC. In a subset of OSCC with discordant p16-positive and HPV DNA ISH-negative status and RB1 loss, we further tested HPV status by HPV RNA ISH and determined the mechanism of RB1 loss by RB1 fluorescence in situ hybridization (FISH) and targeted next generation sequencing including RB1, TP53, p16/CDKN2A, and CDK6. Finally, the prognostic performance of RB1 status was compared to that of p16 IHC and the 7th and 8th edition of AJCC staging.
Material and Methods
Study Population
Two hundred fifty-seven patients with OSCC were identified from 1983 to 2015. Clinical information and demographics were acquired from the Organ Specific Database and Head and Neck Specialized Program of Research Excellence. Tumors were staged according to both the 7th and 8th editions of the AJCC staging classification manual. This work was approved by the Institutional Review Board IRB991206. The study was performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.
Tissue Microarray Construction and Immunohistochemistry (IHC)
Tissue microarrays were built by transferring 2 mm cores from donor FFPE block to a recipient paraffin block using a manual tissue arrayer (Beecher Instruments, Sun Prairie, WI). Hematoxylin and eosin stain, P16 IHC, and RB1 IHC were performed on all 257 cases. For all cases with RB1 loss on TMA, result was corroborated by repeated RB1 IHC on whole tissue sections. Slides were cut at 4 microns thickness and immunohistochemical labeling was performed on the Ventana Benchmark Ultra platform automated stainer (Ventana Medical Systems, Tucson, AZ). Immunohistochemical staining for p16 was performed with pre-diluted antibody clone E6H4 (Ventana, Oro Valley, AZ) and was considered positive when 70% or more of tumor cells showed diffuse and strong nuclear and cytoplasmic staining [4]. Immunohistochemical staining for RB1 was performed with antibody clone 13A10, 1:50 dilution (Leica Biosystems, Buffalo Grove, IL). RB1 expression was considered lost when there was complete absence of nuclear staining in > 90% of tumor cells, in the presence of positive internal control in lymphocytes, endothelial cells, and/or adjacent normal squamous epithelium. Immunohistochemical staining for p53 was performed with DO-7 monoclonal mouse antibody, 1:100 dilution (Dako, Carpinteria, CA). TP53 IHC was performed and interpreted as described previously [17].
HPV DNA In Situ Hybridization and RNA In Situ Hybridization
HPV DNA ISH was performed on all cases, using HPV probe cocktail including strains 6, 11, 16, 18, 31, 33, 35, 39, 45, 51 and 52 (Y1404, Dako, Carpinteria, CA) and punctate nuclear staining was considered positive as previously described [18].
HPV RNA ISH for high risk HPV E6/E7 mRNA was performed on cases with positive p16 and negative HPV DNA ISH (14 of 15 cases; one case had insufficient material). HPV RNA ISH for high risk HPV E6/E7 mRNA was performed using the RNAscope HPV-HR18 probe (Advanced Cell Diagnostics, Hayward, CA), a cocktail probe that recognized 18 HPV types (16, 18, 26, 31, 33, 35, 39, 45, 51, 52, 53, 56, 58, 59, 66, 68, 73, and 82) as previously described [19]. RNA PPIB (cyclophilin B) was used as an external control and positive cases showed punctate nuclear and/or cytoplasmic signals in tumor cells [19].
Retinoblastoma 1 Fluorescence In Situ Hybridization
To identify the mechanism of RB1 loss by IHC in p16-positive OSCC with negative HPV DNA ISH (n = 10), RB1 FISH was performed with locus specific identifier (LSI) 13 (RB1) 13q14 SpectrumOrange Probe (Abbott Molecular, Des Plaines, IL) and control region LSI 13q34 SpectrumGreen Probe (Abbott Molecular, Des Plaines, IL). RB1 FISH was also performed on one case with preserved RB1 expression by IHC, positive p16, and negative HPV DNA and RNA ISH. Images were captured using FISH Imaging System: Leica Biosystems (CytoVision FISH Capture and Analysis Workstation) (Buffalo Grove, IL). Analysis was manually performed and quantitatively assessed on a minimum of 62 cells (range, 62–126). The nuclei were counterstained with DAPI/Antifade 1 (blue) (Vysis, Inc., Downers Grove, IL). The normal pattern of RB1 FISH is characterized by two orange and two green signals per nucleus. RB1 homozygous deletion was defined as > 20% of cells without RB1 locus signal in the presence of ≥ 2 LSI 13q34 signals. RB1 hemizygous deletion was defined as > 30% of cells with only one RB1 locus signal in the presence of ≥ 2 LSI 13q34 signals. Cases with more than 50% of cells with a single LSI 13q34 signal were categorized as Chromosome 13 monosomy. Also, RB1 deletion by FISH may be defined as RB1/LSI < 0.8 and all cases with RB1/LSI < 0.8 in this study also satisfied definition of hemizygous loss as outlined above.
Nucleic Acids Isolation and Targeted Next Generation Sequencing
FFPE tissue from 5 cases with positive p16, negative HPV by DNA ISH, RB1 loss by IHC, and no RB1 alterations by FISH, and one case with positive p16, preserved RB1, and negative HPV by DNA and RNA ISH were sequenced. Nucleic acids were isolated and targeted next generation sequencing (NGS), covering mutations in 30 cancer-associated genes and copy number alterations in 24 cancer-associated genes, and data analysis was performed as previously described [20]. RB1 exons 4, 6, 10, 11, 14, 17–18, 20–22 were covered by NGS panel.
Statistical Analysis
Patients’ characteristics were compared and a t test was used to characterize the relationship between continuous variables and a chi-square test was used to characterize the relationship between categorical variables. Disease free survival (DFS) was assessed from the time of pathologic diagnosis to the time of recurrence (local recurrence or distant metastasis). Living patients were censored at the time of last clinical follow-up or at the time of developing second primary carcinoma of the upper aerodigestive tract. Patients who died of other causes were censored at the time of death. Median disease-free survival intervals with 95% confidence intervals (CIs) were estimated using the Kaplan–Meier method, with statistical significance of differences between groups estimated by the log rank test. For all the tests, a p value of < 0.05 was defined as statistically significant. Statistical analysis was performed using SPSS 19 (Somers, NY).
Results
Patients’ Characteristics and Demographics
Basic clinicopathologic parameters of studied population are summarized in Table 1. Most patients were male, 206/257, 80% and had p16-positive OSCC, 184/257, 72%. Patients with p16-positive OSCC were younger at the time of diagnosis (57 vs. 59.3 years, mean, p < 0.01) and less likely to be smokers (59.2% vs. 73.4%, p < 0.01). p16-positive OSCC more often showed non-keratinizing morphology (88.5% vs. 43.8%, p < 0.01). No statistically significant difference was found between the p16-positive and p16-negative groups in terms of alcohol use and anatomic subsite (Table 1).
Table 1.
Clinicopathologic characteristics of studied patients (n = 257)
| Parameter | P16-negative (n = 73) | P16-positive (n = 184) | p value |
|---|---|---|---|
| Male: female ratio | 2.5:1 | 5.1:1 | 0.02 |
| Age at diagnosis, years (mean, range) | 59.3 (40–85) | 57 (31–78) | < 0.01 |
| Smoking, yes | 54 (73.4%) | 109 (59.2%) | < 0.01 |
| Alcohol use, yes | 40 (54.8%) | 104 (56.5%) | 0.88 |
| Chemotherapya | 19 (27.5%) | 95 (53%) | < 0.01 |
| Radiation therapyb | 43 (60.5%) | 150 (83.8%) | < 0.01 |
| Morphology | |||
| Keratinizing | 34 (46.6%) | 10 (5.4%) | < 0.01 |
| Nonkeratinizing | 32 (43.8%) | 163 (88.5%) | |
| Other | 7 (9.6%) | 11 (6.0%) | |
| Subsite | |||
| Tonsil | 42 (57.5%) | 103 (56.0%) | 0.3 |
| Base of tongue | 21 (28.8%) | 74 (40.2%) | |
| Other | 10 (13.7%) | 7 (3.8%) | |
| pN0 | 23 (31.5%) | 15 (8%) | < 0.01 |
| Stage (AJCC 7th edition)c | |||
| I | 8 (11.0%) | 13 (7.1%) | 0.02 |
| II | 13 (17.8%) | 12 (6.5%) | |
| III | 13 (17.8%) | 35 (19.0%) | |
| IV | 36 (49.3%) | 118 (64.1%) | |
| Stage (AJCC 8th edition)c | |||
| I | 8 (11.0%) | 133 (72.3%) | < 0.01 |
| II | 13 (17.8%) | 39 (21.2%) | |
| III | 13 (17.8%) | 6 (3.3%) | |
| IV | 36 (49.3%) | 0 | |
aChemotherapy details were unknown for 9 patients (4 p16-negative and 5 p16-positive)
bRadiotherapy details were unknown for 6 patients (2 p16-negative and 5 p16-positive)
cStaging information was incomplete for 9 patients (3 p16-negative, and 6 p16-positive)
Clinical Stage Migration for Patients with p16-Positive Oropharyngeal Squamous Cell Carcinoma, 7th and 8th Editions of AJCC Staging Manual
Most patients in the p16-positive group were Stage IV according to the AJCC 7th edition staging classification (Table 1). Under the AJCC 8th edition staging, most patients with p16-positive OSCC were categorized as stage I and none of the p16-positive OSCC patients was classified as Stage IV (Table 2). Overall, 8th edition of AJCC staging resulted in down-staging for patients with p16-positive OSCC (Table 2).
Table 2.
Clinical stage migration for patients with p16-positive oropharyngeal squamous cell carcinoma, 8th and 7th editions of AJCC staging manual
| AJCC 7th edition | AJCC 8th edition | Totala | ||
|---|---|---|---|---|
| I (n = 133) | II (n = 39) | III (n = 6) | ||
| I | 13 | 13 | ||
| II | 12 | 12 | ||
| III | 30 | 4 | 1 | 35 |
| IV | 78 | 35 | 5 | 118 |
aStaging information was incomplete for 6 patients
Disease Free Survival of Patients with Oropharyngeal Squamous Cell Carcinoma
Disease free survival (DFS) was chosen as a primary endpoint because of the number of patients with p16-positive OSCC who died of causes unrelated to primary disease: of 184 patients with p16-positive OSCC, 28 died of disease, while 19 died of other causes. Prognostic performance of the 8th edition of AJCC is shown in Fig. 1a. For instance, the estimated median DFS for patients with Stage III disease was 30 months (95% confidence interval (CI), 5 – 55 months) and 17 months for Stage IV (95% CI 11.8–22.1 months).
Fig. 1.
a Disease free survival of patients with oropharyngeal squamous cell carcinoma (OSCC) staged following 8th edition of American Joint Cancer Committee (AJCC 8th edition) manual. The estimated mean disease-free survival for patients with Stage I disease was 62.7 months (95% confidence interval, 95CI: 59–66 months), for Stage II—56.4 months (95CI 49–63.5 months), Stage III—39.8 months (95CI 26–52.3 months), and Stage IV—32.9 months (95CI 22–43 months). b Disease free survival of patients with OSCC and prognostic performance of p16 immunohistochemistry. The estimated mean disease-free survival for patients with p16-positive OSCC was 59.6 months (95% CI 56–63 months) and for patients with p16-negative OSCC was 42.7 months (95% CI 35.5–50 months). The median disease-free survival for patients with p16-negative or p16-positive SCC was not reached. c and d AJCC 8th edition staging correlated with disease free survival in the sub-group of patients with p16-negative (c) OSCC and p16-positive OSCC (d)
DFS for patients with p16-positive and p16-negative OSCC are shown in Fig. 1b. Clinical staging according to the 8th edition of the AJCC staging manual correlated with DFS in the subgroups of patients with p16-negative OSCC (Fig. 1c) and p16-positive (Fig. 1d).
RB1 Loss by IHC is Seen Only in p16-Positive Oropharyngeal Squamous Cell Carcinomas
The workflow and case selection for and results of ancillary testing are summarized in Fig. 2 and Tables 3 and 4.
Fig. 2.
Summary of the workflow and case selection for study by HPV DNA ISH and HPV RNA ISH, RB1 FISH, and next generation sequencing. Oropharyngeal squamous cell carcinomas with concordant p16 IHC and HPV DNA ISH results are highlighted by green arrows and green boxes. Cases of oropharyngeal squamous cell carcinoma with discordant p16-positive and HPV DNA ISH-negative results (in red) were further studied by HPV RNA ISH (*1 of 15 cases had insufficient material for high risk HPV RNA ISH), RB1 fluorescence in situ hybridization, and next generation sequencing
Table 3.
Relationship between p16 and Retinoblastoma 1 immunohistochemical findings
| p16 IHC results | Retinoblastoma 1 IHC Results | |
|---|---|---|
| Lost (n = 79) | Preserved (n = 178) | |
| Negative (n = 73) | 0 | 73 |
| Positive (n = 184) | 79 | 105 |
IHC immunohistochemistry
Table 4.
Clinicopathologic and molecular data on patients with oropharyngeal squamous cell carcinoma with Retinoblastoma 1 loss by immunohistochemistry and discordant p16-positive and HPV DNA ISH-negative
| Case | Sex, age (years) | Smoker | Morphology | HPV RNA ISH, positive? | RB1 FISH | NGS |
|---|---|---|---|---|---|---|
| 1 | Male, 57 | No | Non-keratinizing | Yes | 56.5% hemizygous deletion | Not done |
| 2 | Male, 62 | Unknown | Non-keratinizing | Yes | 67.2% hemizygous deletion | Not done |
| 3 | Male, 38 | Unknown | Non-keratinizing | Yes | 51.4% monosomy | Not done |
| 4 | Female, 55 | No | Non-keratinizing | Yes | 58.1% monosomy | Not done |
| 5 | Female, 56 | Unknown | Non-keratinizing | Yes | 71.4% monosomy | Not done |
| 6 | Male, 52 | Yes | Non-keratinizing | Yes | Normal | Negative |
| 7 | Female, 68 | No | Non-keratinizing | Yes | Normal | Negative |
| 8 | Male, 53 | Yes | Non-keratinizing | Yes | Normal |
PIK3CA p.H1047R; DDX3X loss |
| 9 | Male, 50 | No | Keratinizing | Yes | Normal | DDX3X loss |
| 10 | Male, 66 | Yes | Keratinizing | No | Normal | DDX3X loss |
None of the p16-negative cases (73/257, 28.4%) showed RB1 loss by IHC (0/73).
Favorable Prognostic Value of RB1 Loss by IHC is p16-Dependent
Loss of RB1 by IHC correlates with better DFS (Fig. 3). Patients with OSCC that showed loss of RB1 had an estimated mean DFS of 60.2 months (95CI 55.4–65.1 months), while patients with OSCC that showed preserved RB1 expression by IHC had an estimated mean DFS of 52.5 months (95CI 48.4 – 56.7 months) (p = 0.041) (Fig. 3a). Since only p16-positive OSCC showed RB1 loss (Table 3), prognostic value of RB1 IHC was tested in a subgroup of patients with p16-positive OSCC. Among patients with p16-positive OSCC RB1 loss by IHC did not correlate with DFS (Fig. 3b).
Fig. 3.
Disease free survival (DFS) of patients with oropharyngeal squamous cell carcinoma (OSCC) and status of Retinoblastoma 1. a Loss of Retinoblastoma 1, by immunohistochemistry, correlates with better disease-free survival (DFS). b Since only p16-positive OSCCs showed loss of Retinoblastoma 1, prognostic value of Retinoblastoma 1 immunohistochemistry was tested in a sub-group of patients with p16-positive OSCC. In a sub-group of patients with p16-positive OSCC, Retinoblastoma 1 loss by immunohistochemistry did not correlate with DFS
HPV DNA ISH-Negative Cases Are Commonly HPV RNA ISH-Positive
The cases with RB1 loss and discordant p16-positive and HPV DNA ISH-negative results were studied further by HPV RNA ISH. HPV RNA ISH was positive in 12 of 14 HPV DNA ISH-negative cases. Overall, most discordant p16-positive and HPV DNA ISH-negative cases of OSCC were ultimately HPV-mediated when tested by HPV RNA ISH.
RB1 IHC Loss as an HPV-Independent Potential Mechanism for p16 Positivity in Oropharyngeal Squamous Cell Carcinomas is Exceedingly Rare
RB1 was lost by IHC in 69 of 169 (40.8%) p16-positive/HPV DNA ISH-positive cases and in 10 of 15 OSCC with discordant p16-positive IHC and HPV DNA ISH-negative. Of the two p16-positive OSCC which were HPV DNA ISH and HPV RNA ISH-negative, only one case showed RB1 loss by IHC (Fig. 4, Table 4, case #10). Therefore, the prevalence of HPV-independent p16 positivity potentially due to RB1 loss is exceedingly low—1/184, 0.5%.
Fig. 4.
Oropharyngeal squamous cell carcinoma with discordant positive p16 IHC, negative HPV DNA in situ hybridization and negative HPV RNA in situ hybridization, case #10 in Table 4. a Invasive squamous cell carcinoma with extracellular keratinization, H&E, 200x. b Positive p16, immunohistochemistry, 100x. c Retinoblastoma 1 loss in the invasive carcinoma and preserved nuclear Retinoblastoma 1 expression in stromal cells and intratumoral lymphocytes, immunohistochemistry, 400x. Note, there were no Retinoblastoma 1 alterations by fluorescent in situ hybridization or Retinoblastoma 1 sequencing
The only p16-positive/HPV DNA ISH and HPV RNA ISH-negative OSCC with RB1 loss by IHC affected a 66 year old man, a smoker. DDX3X loss was identified by NGS. This patient underwent adjuvant chemoradiotherapy and was alive and disease free at 34 months of follow up.
The second case of p16-positive and HPV DNA and HPV RNA ISH-negative OSCC showed preserved RB1 expression by IHC and normal RB1 by FISH but revealed telomerase reverse transcriptase (TERT) C228T and TP53 p.K132E mutations. TP53 mutation was corroborated by IHC with clonal (strong and diffuse) p53 expression found in adjacent squamous dysplastic mucosa and invasive carcinoma. This patient was a 66 year old female, smoker, with tonsillar keratinizing squamous cell carcinoma. The patient did not receive adjuvant therapy and was lost to follow up.
RB1 Hemizygous Deletion or Chromosome 13 Monosomy by FISH Were Identified in Five of Ten Oropharyngeal Squamous Cell Carcinomas with Discordant p16-Positive/HPV DNA ISH-Negative with RB1 Loss by IHC
All OSCC with discordant p16-positive/HPV DNA ISH-negative with RB1 IHC loss (n = 10) were studied by RB1 FISH. Three OSCCs showed 13q34 monosomy and 2 revealed hemizygous deletion (Table 4 and Fig. 5). The remaining 5 cases with discordant p16-positive/HPV DNA ISH-negative results with RB1 loss by IHC did not demonstrate RB1 alterations by sequencing or FISH (Table 4).
Fig. 5.
Representative case of an oropharyngeal squamous cell carcinoma with discordant p16-positive, HPV DNA in situ hybridization (ISH) negative, HPV RNA ISH-positive, and Retinoblastoma 1 loss by immunohistochemistry with hemizygous deletion of Retinoblastoma 1 (case #2 in Table 4). a Invasive oropharyngeal squamous cell carcinoma with overlying normal squamous mucosa, H&E, 200X, b Retinoblastoma 1 is lost in the invasive carcinoma component, while Retinoblastoma 1 expression is preserved in internal positive controls, i.e., overlying normal squamous mucosa and lymphocytes, immunohistochemistry, 200x. c Positive p16 in the invasive carcinomatous component and negative p16 in overlying normal squamous mucosa, immunohistochemistry, 200x. d Hemizygous deletion of Retinoblastoma 1 with 67.2% of cells showing one Retinoblastoma 1 signal (orange) while maintaining two 13q34 signals (green), fluorescent in situ hybridization, 600x
Co-existing Molecular Alterations Identified by NGS
Five cases of p16-positive/HPV DNA ISH-negative/RB1 lost by IHC and with normal RB1 by FISH OSCC were further tested for mutations by targeted NGS panel covering RB1 exons 4, 6, 10, 11, 14, 17–18, and 20–22, along with TP53, CDKN2A, CDK6. No RB1, TP53, CDKN2A, or CDK6 mutations or copy number alterations were identified, while DDX3X deletion was found in 3 cases. In one case (case #8, Table 4) PIK3CA p.H1047R mutation was identified.
Discussion
P16 IHC is an established surrogate marker of HPV in OSCC [4]. The knowledge of HPV status in OSCC is essential as patients with p16-positive OSCC have better outcome and are staged differently as per 8th edition of AJCC staging manual. In our cohort, AJCC 8th edition staging system correlates with DFS, overall and in p16-positive and p16-negative subgroups [21].
In the absence of HPV, p16 overexpression can be seen in tumors with RB1 alterations [12, 13]. Prevalence of RB1 loss and RB1 genetic alterations in HPV-negative OSCC with p16 overexpression have not been well established. In prior studies relying on non-ISH based HPV detection (i.e., PCR), the proportion of p16-positive/RB1 lost/HPV-negative OSCC varied from 7.1 to 6.6% [5, 11]. Perhaps even more diagnostically relevant in the setting of metastatic SCCUP, it was previously reported that 25% of cutaneous squamous cell carcinomas are characterized by a p16-positive /RB1 lost immunoprofile [13]. We show that RB1 loss in OSCC with discrepant p16 overexpression and negative HPV by two different HPV detection methods (HPV DNA and RNA ISH) is extremely rare. Indeed, of 184 p16-positive OSCC, only one OSCC showed RB1 loss while being HPV DNA ISH and HPV RNA ISH-negative. Also, while this OSCC showed RB1 loss by IHC, it did not show p16/CDKN2A, CDK6, or RB1 copy number alterations (by FISH) or mutations, suggesting that another mechanism, such as an epigenetic RB1 modification (e.g., promoter methylation), may have led to loss of RB1 [13].
In our study, we found that RB1 loss is exclusively seen in p16-positive OSCC, most of which are HPV-mediated, indicating that p16 overexpression in this setting is secondary to RB1 inactivation by HPV oncoprotein E7 [6]. This is consistent with previously reported associations between p16 positivity as a surrogate marker of HPV and RB1 loss [5]. In addition to HPV E7 oncoprotein induced RB1 degradation, p16-positive HPV-mediated OSCC also revealed RB1 hemizygous deletion or monosomy, as demonstrated by RB1 FISH.
RB1 expression in OSCC has been correlated with a favorable outcome in some studies [5, 11, 14, 15]. However, favorable prognostic significance of RB1 loss was not supported by a study of TCGA and FoundationOne databases [16]. In our cohort, RB1 loss by IHC correlates with better DFS. However, when only p16-positive OSCC are considered, its prognostic significance is lost. This is explained by the fact that RB1 loss occurs exclusively in p16-positive OSCC. Therefore, while RB1 loss by IHC is a favorable prognosticator, it is p16-dependent and offers no additional prognostic information once p16-positivity is established.
The combination of HPV RNA ISH-positivity and HPV DNA ISH-negativity is usually seen in OSCCs with low to intermediate HPV viral copy numbers [22]. It is also possible that the two p16-positive and HPV RNA ISH-negative cases carry an HPV subtype that is not covered by the broad cocktail probe used here. In our study, co-occurring molecular alterations in p16-positive OSCC with apparent low viral load (i.e., HPV DNA ISH-negative and HPV RNA ISH-positive) have been found—TERT, PIK3CA, TP53 mutations, and DDX3X loss. The great majority of HPV-mediated base of tongue carcinomas have been reported to have wild-type TERT promoter [23]. C228T TERT mutation has been more frequently reported in laryngeal tumors in smokers and is associated with poor outcome [23]. TP53 mutations have been associated with poor prognosis in HPV-negative OSCC [24]. It has been found that TP53 mutations were associated with decreased overall survival (OS) in head and neck squamous cell carcinomas, including oropharyngeal carcinomas, regardless of HPV status [23, 25]. It is unclear whether keratinizing morphology and lower HPV viral load are associated with increased prevalence of TERT and TP53 mutations in p16-positive OSCC.
The only p16-positive/HPV DNA and RNA ISH-negative OSCC with RB1 loss showed DDX3X loss. DDX3X loss has been associated with worse outcome in non-smokers with oral squamous cell carcinomas [26]. Interestingly, a study by Seiwert et al. reported DDX3X mutations exclusively in HPV-positive head and neck squamous cell carcinomas [27]. Here we report a p16-positive/HPV-negative keratinizing OSCC with RB1 loss and co-occurring DDX3X loss. This mutational profile appears to be extremely rare and its significance remains uncertain.
This study covers an about 3-decade time period. Inclusion of OSCC cases from 1980s allowed to identify more p16-negative OSCC and outcome reflects treatment patterns before the introduction of treatment de-intensification clinical trials. Of various methods of HPV detection, this study employed the most practical and widely used ones by pathologists—p16 IHC, HPV DNA ISH and HPV RNA ISH, the latter being the most accurate and specific indicator of oncologically relevant and transcriptionally active HPV infection [4, 22].
In conclusion, this retrospective study of 257 patients with OSCC confirms that the 8th edition of the AJCC staging manual satisfactorily stratifies patients with OSCC according to outcome, specifically DFS, overall and in p16-positive and p16-negative subgroups. The adequacy of p16 IHC as a surrogate HPV biomarker in OSCC and superiority of HPV RNA ISH over HPV DNA ISH is confirmed. It unclear whether TERT and TP53 mutations are more common in p16-positive OSCC with apparently lower HPV viral loads (HPV DNA ISH-negative and HPV RNA ISH-positive). RB1 loss by IHC is a favorable p16-dependent prognosticator. RB1 loss in p16-positive /HPV-negative OSCC is extremely rare and represents a negligible and significantly smaller than previously suggested source of potential false positive p16 IHC result.
Funding
No funding obtained.
Declarations
Conflict of interest
No conflict of interest to disclose.
Ethical Approval
All tests performed in this retrospective study involving human participants were in accordance with the ethical standards of the institutional review board (IRB991206), which did not require informed consent.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Ang KK, Harris J, Wheeler R, Weber R, Rosenthal DI, Nguyen-Tan PF, et al. Human papillomavirus and survival of patients with oropharyngeal cancer. N Engl J Med. 2010;363(1):24–35. doi: 10.1056/NEJMoa0912217. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Fakhry C, Westra WH, Li S, Cmelak A, Ridge JA, Pinto H, et al. Improved survival of patients with human papillomavirus-positive head and neck squamous cell carcinoma in a prospective clinical trial. J Natl Cancer Inst. 2008;100(4):261–269. doi: 10.1093/jnci/djn011. [DOI] [PubMed] [Google Scholar]
- 3.Amin MBE, S.B. Greene, F. et al. AJCC Cancer Staging Manual 8th ed. New York, NY: Springer. 2017.
- 4.Lewis JS, Jr, Beadle B, Bishop JA, Chernock RD, Colasacco C, Lacchetti C, et al. Human papillomavirus testing in head and neck carcinomas: guideline from the college of american pathologists. Arch Pathol Lab Med. 2018;142(5):559–597. doi: 10.5858/arpa.2017-0286-CP. [DOI] [PubMed] [Google Scholar]
- 5.Holzinger D, Flechtenmacher C, Henfling N, Kaden I, Grabe N, Lahrmann B, et al. Identification of oropharyngeal squamous cell carcinomas with active HPV16 involvement by immunohistochemical analysis of the retinoblastoma protein pathway. Int J Cancer. 2013;133(6):1389–1399. doi: 10.1002/ijc.28142. [DOI] [PubMed] [Google Scholar]
- 6.Boyer SN, Wazer DE, Band V. E7 protein of human papilloma virus-16 induces degradation of retinoblastoma protein through the ubiquitin-proteasome pathway. Can Res. 1996;56(20):4620–4624. [PubMed] [Google Scholar]
- 7.Rocco JW, Sidransky D. p16(MTS-1/CDKN2/INK4a) in cancer progression. Exp Cell Res. 2001;264(1):42–55. doi: 10.1006/excr.2000.5149. [DOI] [PubMed] [Google Scholar]
- 8.Scheffner M, Werness BA, Huibregtse JM, Levine AJ, Howley PM. The E6 oncoprotein encoded by human papillomavirus types 16 and 18 promotes the degradation of p53. Cell. 1990;63(6):1129–1136. doi: 10.1016/0092-8674(90)90409-8. [DOI] [PubMed] [Google Scholar]
- 9.Lechner M, Chakravarthy AR, Walter V, Masterson L, Feber A, Jay A, et al. Frequent HPV-independent p16/INK4A overexpression in head and neck cancer. Oral Oncol. 2018;83:32–37. doi: 10.1016/j.oraloncology.2018.06.006. [DOI] [PubMed] [Google Scholar]
- 10.Li Y, Nichols MA, Shay JW, Xiong Y. Transcriptional repression of the D-type cyclin-dependent kinase inhibitor p16 by the retinoblastoma susceptibility gene product pRb. Can Res. 1994;54(23):6078–6082. [PubMed] [Google Scholar]
- 11.Plath M, Broglie MA, Forbs D, Stoeckli SJ, Jochum W. Prognostic significance of cell cycle-associated proteins p16, pRB, cyclin D1 and p53 in resected oropharyngeal carcinoma. J Otolaryngol Head Neck Surg. 2018;47(1):53. doi: 10.1186/s40463-018-0298-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Svajdler M, Mezencev R, Ondic O, Saskova B, Mukensnabl P, Michal M. P16 is a useful supplemental diagnostic marker of pulmonary small cell carcinoma in small biopsies and cytology specimens. Ann Diagn Pathol. 2018;33:23–29. doi: 10.1016/j.anndiagpath.2017.11.008. [DOI] [PubMed] [Google Scholar]
- 13.Murao K, Kubo Y, Ohtani N, Hara E, Arase S. Epigenetic abnormalities in cutaneous squamous cell carcinomas: frequent inactivation of the RB1/p16 and p53 pathways. Br J Dermatol. 2006;155(5):999–1005. doi: 10.1111/j.1365-2133.2006.07487.x. [DOI] [PubMed] [Google Scholar]
- 14.Ryu CH, Ryu J, Cho KH, Moon SH, Yun T, Lee SH, et al. Human papillomavirus-related cell cycle markers can predict survival outcomes following a transoral lateral oropharyngectomy for tonsillar squamous cell carcinoma. J Surg Oncol. 2014;110(4):393–399. doi: 10.1002/jso.23672. [DOI] [PubMed] [Google Scholar]
- 15.Andl T, Kahn T, Pfuhl A, Nicola T, Erber R, Conradt C, et al. Etiological involvement of oncogenic human papillomavirus in tonsillar squamous cell carcinomas lacking retinoblastoma cell cycle control. Can Res. 1998;58(1):5–13. [PubMed] [Google Scholar]
- 16.Beck TN, Smith CH, Flieder DB, Galloway TJ, Ridge JA, Golemis EA, et al. Head and neck squamous cell carcinoma: Ambiguous human papillomavirus status, elevated p16, and deleted retinoblastoma 1. Head Neck. 2017;39(3):E34–E39. doi: 10.1002/hed.24604. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Chiosea SI, Williams L, Griffith CC, Thompson LD, Weinreb I, Bauman JE, et al. Molecular characterization of apocrine salivary duct carcinoma. Am J Surg Pathol. 2015;39(6):744–752. doi: 10.1097/PAS.0000000000000410. [DOI] [PubMed] [Google Scholar]
- 18.Chenevert J, Chiosea S. Incidence of human papillomavirus in oropharyngeal squamous cell carcinomas: now and 50 years ago. Hum Pathol. 2012;43(1):17–22. doi: 10.1016/j.humpath.2011.03.009. [DOI] [PubMed] [Google Scholar]
- 19.Rooper LM, Gandhi M, Bishop JA, Westra WH. RNA in-situ hybridization is a practical and effective method for determining HPV status of oropharyngeal squamous cell carcinoma including discordant cases that are p16 positive by immunohistochemistry but HPV negative by DNA in-situ hybridization. Oral Oncol. 2016;55:11–16. doi: 10.1016/j.oraloncology.2016.02.008. [DOI] [PubMed] [Google Scholar]
- 20.Nikiforova MN, Wald AI, Melan MA, Roy S, Zhong S, Hamilton RL, et al. Targeted next-generation sequencing panel (GlioSeq) provides comprehensive genetic profiling of central nervous system tumors. Neuro Oncol. 2016;18(3):379–387. doi: 10.1093/neuonc/nov289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Lydiatt WM, Patel SG, O'Sullivan B, Brandwein MS, Ridge JA, Migliacci JC et al. Head and Neck cancers-major changes in the American Joint Committee on cancer eighth edition cancer staging manual. CA Cancer J Clin. 2017;67(2):122–37. doi:10.3322/caac.21389. [DOI] [PubMed]
- 22.Keung ES, Souers RJ, Bridge JA, Faquin WC, Graham RP, Hameed MR, et al. Comparative performance of high-risk human papillomavirus RNA and DNA in situ hybridization on college of American pathologists proficiency tests. Arch Pathol Lab Med. 2020;144(3):344–349. doi: 10.5858/arpa.2019-0093-CP. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Boscolo-Rizzo P, Da Mosto MC, Rampazzo E, Giunco S, Del Mistro A, Menegaldo A, et al. Telomeres and telomerase in head and neck squamous cell carcinoma: from pathogenesis to clinical implications. Cancer Metastasis Rev. 2016;35(3):457–474. doi: 10.1007/s10555-016-9633-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Resteghini C, Perrone F, Miceli R, Bergamini C, Alfieri S, Orlandi E, et al. Prognostic role of PIK3CA and TP53 in human papillomavirus-negative oropharyngeal cancers. Tumori. 2018 doi: 10.1177/0300891618765558. [DOI] [PubMed] [Google Scholar]
- 25.Wichmann G, Rosolowski M, Krohn K, Kreuz M, Boehm A, Reiche A, et al. The role of HPV RNA transcription, immune response-related gene expression and disruptive TP53 mutations in diagnostic and prognostic profiling of head and neck cancer. Int J Cancer. 2015;137(12):2846–2857. doi: 10.1002/ijc.29649. [DOI] [PubMed] [Google Scholar]
- 26.Lee CH, Lin SH, Yang SF, Yang SM, Chen MK, Lee H, et al. Low/negative expression of DDX3 might predict poor prognosis in non-smoker patients with oral cancer. Oral Dis. 2014;20(1):76–83. doi: 10.1111/odi.12076. [DOI] [PubMed] [Google Scholar]
- 27.Seiwert TY, Zuo Z, Keck MK, Khattri A, Pedamallu CS, Stricker T, et al. Integrative and comparative genomic analysis of HPV-positive and HPV-negative head and neck squamous cell carcinomas. Clin Cancer Res. 2015;21(3):632–641. doi: 10.1158/1078-0432.Ccr-13-3310. [DOI] [PMC free article] [PubMed] [Google Scholar]