PDF
Citation
Print
Head and neck cancer includes all tumors that arise from various anatomical subsites, including the oral cavity, oropharynx, nasopharynx, hypopharynx, larynx, paranasal sinuses, and salivary glands. Although histologic subtypes are diverse, malignancies of the epithelial lining, namely squamous cell carcinoma, are the most common [1].
Head and neck squamous cell carcinoma (HNSCC) is the sixth most common malignancy worldwide, with approximately 890,000 new cases diagnosed and 450,000 deaths reported in 2008. Among all HNSCC, the oral cavity is the most common subsite of a primary tumor [2]. The burden of HNSCC varies across regions of the world. In Thailand, HNSCC is one of the most common cancers, with approximately 10,000 new cases diagnosed annually. The age-standardized incidence rates of HNSCC in males and females are 15.7 and 10.7, respectively [3]. The dominant tumor locations of HNSCC in males are the oral cavity, nasopharynx, and larynx; and the most common location in females is the oral cavity [3].
At the time of diagnosis, most patients with HNSCC have locally advanced, unresectable tumors. Despite advanced multidisciplinary treatments, including surgery with either or both radiotherapy and chemotherapy, only modest improvements in outcomes have been achieved over the past 20 years. HNSCC molecular profiling and biomarker studies have provided information regarding rapid diagnosis, prognosis, and surveillance for recurrence and metastasis. However, biomarkers that serve as potential targets of novel therapies to improve survival outcomes are important. Among the signaling pathways associated with tumor proliferation and survival, the phosphoinositide 3-kinase/AKT/mammalian target of rapamycin (PI3K/AKT/mTOR) pathway is the most frequently altered oncogenic pathway in HNSCC [4,5].
The phosphatidylinositol 3-kinase gene (PI3K) encodes PI3K enzymes (PI3Ks), which ultimately trigger various downstream signaling pathways that result in cell survival, apoptosis, transformation, metastasis, and cell migration. PI3Ks are classified into three major subclasses (class I, II, and III) that are determined based on PI3K structure and substrate specificity. Phosphatidylinositol- 4,5-bisphosphate 3-kinase catalytic subunit alpha (PIK3CA) is a 34-kb gene located on chromosome 3q26.3 that consists of 20 exons coding for 1,068 amino acids, yielding a protein with a molecular weight of 124 kDa (p110α catalytic subunit protein, class I) [6]. Several studies previously reported alterations in PIK3CA in many cancers. The presence of PIK3CA mutations is associated with prognosis and might predict response to PI3K inhibitors [6,7]. In HNSCC, the frequency of PIK3CA mutations varies among primary tumor locations [6,8–15]; hence, the frequency of these mutations from different tumor subsites might be relevant to determine treatments and outcomes among HNSCCs. Furthermore, the mutation rate in PIK3CA might be diverse among ethnicities [15,16]; the mutation rate is unknown in Southeast Asian populations, including the Thai population. Therefore, in the present study, we evaluated the frequency of PIK3CA mutations in patients with squamous cell carcinoma of the oral cavity, oropharynx, and hypopharynx in southern Thailand. Furthermore, the clinical characteristics and survival between patients with or without PIK3CA mutations were analyzed and compared.
MATERIALS AND METHODS
Patients and specimens
During 2002–2003, 96 fresh tissues (63 from the oral cavity, 10 from the oropharynx, and 23 from the hypopharynx) were obtained from patients registered at the head and neck clinic of Songklanagarind Hospital. The tissue samples were obtained from a biopsy or sampling of surgical resection specimen, after which they were snap frozen and stored at −80°C until DNA extraction. All collected tissues were subsequently reviewed to ensure cancer site and pathological diagnosis of cancer by a pathologist (P.T). All cases were primary tumors that had not been treated with radiation or chemotherapy.
Clinicopathological parameters of the patients were retrieved from medical records. The patient information extracted from the database was date of birth, sex, vital status, history of smoking and alcohol consumption, date of diagnosis, stage, and treatment. Pathological information was obtained from pathological reports. Date of death was obtained from the national civil registration system.
Mutation detection using conventional polymerase chain reaction and DNA sequencing
Genomic DNA was extracted from fresh tissue using an E.Z.N.A Tissue DNA Kit (Omega Bio-Tek Inc., Norcross, GA, USA), and the procedures were performed according to the manufacturer’s instructions. PIK3CA gene analysis was performed in exons 4, 9, and 20; 100 ng of DNA was amplified using polymerase chain reaction (PCR) with the following primer sets: PIK-3CA-exon 4 forward (5′-CATCTTATTCCAGACGCATTTC-3′), PIK3CA-exon 4 reverse (5′-AGATTACTGTATAGTGCAAGA AAA-3′), PIK3CA-exon 9 forward (5′-CCAGAGGGGAAAAA TATGACA-3′), PIK3CA-exon 9 reverse (5′-CATTTTAGCACT TACCTGTGAC-3′), PIK3CA-exon 20 forward (5′-CATTTGC TCCAAACTGACCA-3′), and PIK3CA-exon 20 reverse (5′-TG AGCTTTCATTTTCTCAGTTATCTTTTC-3′). The PCR conditions for PIK3CA exons 4 and 9 were denaturation at 95°C for 3 minutes; amplification for 45 cycles at 95°C for 30 seconds, 58°C for 30 seconds, and 72°C for 45 seconds; and extension at 72°C for 7 minutes. The PCR conditions for PIK3CA exon 20 were denaturation at 95°C for 3 minutes; amplification for 45 cycles at 95°C for 45 seconds, 58°C for 45 seconds, and 72°C for 1 minute; and extension at 72°C for 7 minutes. All gene sequencing was performed with an ABI PRISM 3730xl Genetic Analyzer (Applied Biosystems, Foster City, CA, USA). DNA sequences were analyzed to ascertain point mutations in PIK3CA exons 4, 9, and 20. The conventional PCR technique was performed in certain samples to determine the reference point mutations. Then, allele-specific real-time PCR was performed for the remaining tissue samples.
Allele-specific real-time PCR
Reference point mutations were used to design allele-specific probes (dual-labeled fluorogenic probes) to develop a real-time PCR technique. The allele-specific probes were composed of short DNA sequences covering the mutational areas. The probes were labeled with two fluorescent dyes: HEX for the PIK3CA wild-type allele and FAM for the mutated allele (Sigma-Aldrich, Singapore). BHQ-1 was used as a quencher. For all samples, 20 ng of DNA was obtained and subjected to real-time PCR using previously described primer sets and allele-specific probe sets (Table 1). PCR mixtures contained 10 μL of 2 × FastStart Essential DNA Probes Master buffer (Roche Diagnostics, Mannheim, Germany), 0.5 μL forward primer, 0.5 μL reverse primer, 0.25 μL wild-type probe, 0.25 μL mutated probe, 6.5 μL molecular grade water, and 20 ng DNA in a final volume of 20 μL. The real-time PCR conditions for PIK3CA exon 9 were 95°C, 3 minutes and (95°C, 10 seconds; 58°C, 30 seconds) × 45 cycles. The real-time PCR conditions for PIK3CA exon 20 were 95°C, 3 minutes and (95°C, 10 seconds; 62°C, 30 seconds; 72°C, 30 seconds) × 45 cycles. Template controls were not included in each real-time PCR run. All samples were subsequently subjected to real-time PCR using the probes. Five mutant specimens served as positive controls.
Statistical analysis
Descriptive statistics were used to compare the data between the two groups of patients. The chi-square test or Fisher exact test was used to compare proportional data, and Student’s t test or the Mann-Whitney U test was used to assess the differences in continuous data. Descriptive results are presented as the median (continuous variable) and percentage (categorical variable).
Duration of follow-up and overall survival (OS) time were calculated from the date of diagnosis until the date of last follow-up or date of death from any cause, respectively. The data of patients who were alive at the last date of follow-up, December 20, 2021, were censored. The survival curve was calculated using the Kaplan-Meier method. Comparisons of survival curve among groups were performed using the log-rank test.
Stata/MP ver. 17.0 (StataCorp, College Station, TX, USA) was used to perform all analyses. p-values < .05 were considered statistically significant.
RESULTS
The samples used in this study comprised 96 HNSCC samples diagnosed from 2002–2003. The study included primary tumors originating from the oral cavity (n = 63), hypopharynx (n = 23), and oropharynx (n = 10). The patient demographics and their baseline clinical characteristics were available for 84 cases and are shown in Table 2; 49 cases were male, and the median age at diagnosis was 68 years.
We initially identified three PIK3CA point mutations using conventional PCR (Fig. 1A–C): one in exon 9 and two in exon 20. Mutations were not detected in exon 4. We then applied these known mutations as a reference to design allele-specific probes (dual-labeled fluorogenic probes) as shown in Table 1.
We found PIK3CA mutations in 10 of 96 samples (10.4%) with all heterozygous mutations. When the patients were classified based on mutational events in the primary tumor location, 17.4% (4/23) with hypopharyngeal squamous cell carcinoma (HPSCC) and 9.5% (6/63) with oral cavity squamous cell carcinoma (OCSCC) had PIK3CA mutations. No patient (0/10) with oropharyngeal squamous cell carcinoma (OPSCC) had PIK3CA mutations (Table 3). Clinical characteristics between patients with or without PIK3CA mutations were not statistically significantly different (Table 2). The patient characteristics and details of PIK3CA mutations are shown in Table 4. Two patients had exon 9 mutations (Fig. 2A) and eight had exon 20 mutations (Fig. 2B, C). The 10 mutations were categorized as five missense and five silent mutations, which corresponded to E545K in exon 9 and T1025T and H1047R in exon 20. E545K (GAG1633AAG) and H1047R (CAT3140CGT) are missense mutations, and T1025T (ACC3075ACT) is a silent mutation.
The majority of patients with wild-type PIK3CA (51.4%) underwent surgery followed by adjuvant radiotherapy; however, patients with PIK3CA mutations underwent only surgery (50.0%), as shown in Table 2. The median duration of follow-up was 2.11 years (95% confidence interval [CI], 1.21 to 3.26). The 5-year OS rates in the wild-type PIK3CA and mutant PIK3CA were 35% (95% CI, 24 to 46) and 58% (95% CI, 23 to 82), respectively (Fig. 3). There was no statistically significant difference in OS rate between the two groups (p = .625).
DISCUSSION
In HNSCC, the PI3K/AKT/mTOR pathway is the most frequently altered oncogenic pathway associated with tumor proliferation and survival [4,5]. Reportedly, PIK3CA mutations are associated with advanced stage of HNSCC, particularly in subjects with oral squamous cell carcinoma, indicating that PIK-3CA mutations play an important role in the carcinogenesis of HNSCC [9,15]. In the present study, we evaluated the frequency of PIK3CA mutations in the different subsites of HNSCC in the Thai population.
When divided by primary tumor location, we detected a higher mutation rate in HPSCC than in OCSCC (17.4% vs. 9.5%). We also found a slightly higher PIK3CA mutation rate in HPSCC than reported in a study in the Chinese population (13.6%) [16]. The frequency of PIK3CA mutations in OCSCC in the present study (9.5%) was slightly higher than the 7.4% frequency reported in the Japanese population [9]. Notably, Kommineni et al. [15] reported a substantially high PIK3CA mutation rate of 52% in OCSCC tumor samples from Indian patients. Taken together, these findings might highlight the effect of diverse mutation rates even among Asian populations.
Human papillomavirus (HPV) infection is a well-known risk factor associated with OPSCC. Nichols et al. [17] reported a higher frequency of PIK3CA mutations in HPV-related OPSCC than HPV-negative tumors (28% vs. 10%). In the present study, we found no PIK3CA mutation in any patient with OPSCC. This result was discordant from those of a study by Beaty et al. [18] showing a PIK3CA mutation rate of 21% in patients with HPV-associated OPSCC. The reason for the discordance might be the small sample cohort in the OPSCC subgroup in the present study. Furthermore, information regarding HPV status in these patients was unknown due to the lack of sufficient specimens. However, in previous studies, the proportion of HPV-related OPSCC in Thai patients was remarkably low (15%–20%) [19,20] compared with 40%–60% reported in Western patients [21]. Thus, the correlation between PIK3CA mutations and HPV status should be further explored in Asian populations.
In the present study, PIK3CA mutations in exon 4 were not detected. In contrast, a previous study reported mutations in exon 4, with a rate of 2.6% in an oral cavity carcinoma cell line (primary tumor originated in the tongue) [8]. In the present study, the overall 10.4% mutation rate of PIK3CA, which included 2.1% in exon 9 (E545K) and 8.3% in exon 20 (H1047R and T1025T), was not significantly different from previous studies (9%–13% in non-nasopharyngeal carcinoma), as shown in Table 5. However, a higher mutation rate of 21% in pharyngeal carcinoma and 21% in oropharyngeal carcinoma were reported in two studies [11,18]. In those studies, a more advantageous technique, a novel mutant-enriched sequencing method, was used for analysis, which could explain the difference in results.
We found two of the three types of PIK3CA mutations, E545K and H1047R, which have been previously reported as hotspot mutations [22–24]. E545K is located on exon 9 in the helical domain of PIK3CA, and H1047R and T1025T are encoded by exon 20 within the kinase domain [25]. The E545K mutation disrupts an inhibitory charge–charge interaction between p110α and the N-terminal SH2 domain of the p85 regulatory subunit, and the H1047R mutation increases the binding affinity between p110α and the negatively charged phosphatidylinositol substrate [26–28]. The functional effect of these two hotspot mutations has been well established in many studies, showing the gain-of-function PI3K activity and downstream signaling via the AKT/mTOR pathway [22,23]. Furthermore, the benefit of PIK3CA mutations as a predictive marker for novel targeted therapies has been demonstrated. Alpelisib, an oral selective PI3K inhibitor, has been implemented in current practice for patients with advanced breast cancer harboring PIK3CA mutations [29,30]. The efficacy and safety of PI3K or mTOR inhibitors are being evaluated in several clinical trials for patients with advanced HNSCC [31]. In contrast to the previously mentioned two hotspot mutations, T1025T is categorized as a silent mutation. Notably, we detected the T1025T mutation, which was significantly high in OCSCC, in 4 of 5 mutant cases in the present study. This was concordant with a previous study in which the T1025T mutation was reportedly prominent in OCSCC [15]. However, whether the T1025T silent mutation is the oncogenic driver in HNSCC remains controversial. Kommineni et al. [15] reported a significantly higher frequency of T1025T silent mutation in patients with HNSCC (41.0%) than in healthy controls (16.7%, p < .001). In contrast, Dogruluk et al. [28] identified T1025T as a genetic variant in the human germline.
In the present study, clinical significance was not observed based on the presence or absence of PIK3CA mutations. This finding was similar to a previous study in which significant difference was not observed in clinical characteristics between patients with wild-type PIK3CA or mutant PIK3CA OPSCC [18]. However, the presence of a PIK3CA mutation in that study was associated with poorer disease-free survival among patients receiving de-intensified definitive chemoradiotherapy for HPV-related OPSCC [18]; survival did not differ between the two groups in the present study. This contradictory finding might be explained by the small sample cohort in the present study. A larger study in which PIK3CA mutations are investigated as a potentially prognostic factor in patients with OPSCC and non-OPSCC is warranted.
To the best of our knowledge, this is the first study to report the frequency of PIK3CA mutations in HNSCC in Thailand. However, the limitations of the study included small sample size, particularly in the OPSCC subgroup, and lack of available data regarding HPV status. The absence of PIK3CA mutations in OPSCC might be due to the smaller number of samples tested than those at other subsites. Furthermore, testing for PIK3CA mutations in normal tissue from these patients was not performed; hence, conclusions cannot be drawn regarding involvement of the T1025T silent mutation in the pathogenesis of HNSCC or its involvement as a genetic variant in the human germline.
In conclusion, our study showed a higher frequency of PIK-3CA mutations in HPSCC than in OCSCC. The results suggest that these mutations are involved in carcinogenesis and could serve as biomarkers for novel targeted therapies as well as immunomodulators in the era of immune checkpoint inhibitors. Furthermore, the hypopharynx is suggested as a primary site of interest for further studies, particularly in Southeast Asian populations.
XML Download