Abstract
Phosphatidylinositol 3‐kinases (PI3K) are a group of heterodimeric lipid kinases that regulate many cellular processes. Gene amplification and somatic mutations mainly within the helical (exon 9) and kinase (exon 20) domains of PIK3CA, which encode the 110‐kDa catalytic subunit of PI3K and are mapped to 3q26, have been reported in various human cancers. Herein, 14 human oral squamous cell carcinoma (OSCC) cell lines and 108 primary OSCC tumors were investigated for activating mutations at exons 9 and 20 as well as amplifications in PIK3CA. PIK3CA missense mutations in exons 9 and 20 were identified in 21.4% (3/14) of OSCC cell lines and 7.4% (8/108) of OSCC tumors by genomic DNA sequencing. An increase in the copy number of PIK3CA, although small, was detected in 57.1% (8/14) of OSCC lines and 16.7% (18/108) of OSCC tumors using quantitative real‐time PCR. A significant correlation between somatic mutations of PIK3CA and disease stage was observed: the frequency of mutations was higher in stage IV (16.1%, 5/31) than in a subset of early stages (stages I–III) (3.9%, 3/77; P = 0.042, Fisher's extract test). In contrast, the amplification of PIK3CA was observed at a similar frequency among all stages. AKT was highly phosphorylated in OSCC cell lines with PIK3CA mutations compared to those without mutations, despite the amplification. The results suggest that somatic mutations of the PIK3CA gene are likely to occur late in the development of OSCC, and play a crucial role through the PI3K–AKT signaling pathway in cancer progression. (Cancer Sci 2006; 97: 1351–1358)
The phosphatidylinositol 3‐kinase (PI3K) signaling pathway is a crucial regulator of many normal cellular processes, such as cell growth, proliferation, motility, survival and apoptosis, and is deregulated in a wide range of human cancers by gain‐ or loss‐of‐function of several components of this pathway, including PTEN, AKT and PIK3CA.( 1 , 2 , 3 ) PIK3CA, a key element of the PI3K–AKT pathway, is located in chromosomal region 3q26.3, and encodes the 110‐kDa catalytic subunit of class IA PI3K. Although an increased DNA copy‐number of PIK3CA is frequently found in tumors, somatic mutations in PIK3CA were recently identified at significant frequencies in various types of human cancer.( 4 , 5 ) More than 80% of these mutations are clustered in the helical domain encoded by exon 9 and the kinase domain encoded by exon 20. In addition, three hotspot mutations in these exons, E542K, E545K and H1047R, were proven to activate the PI3K–AKT pathway through the phosphorylation of AKT and result in transformation in vitro.( 6 , 7 , 8 , 9 ) Such evidence demonstrates that oncogenic activation of the PI3K–AKT pathway through the mutated PIK3CA is involved in cancer development.
Oral squamous cell carcinoma (OSCC) is a subset of head and neck squamous cell carcinoma (HNSCC) involving the oral cavity, pharynx, and larynx. In Japan, OSCC is relatively common, accounting for more than 5500 deaths in 2003.( 10 ) Oncogenesis is generally considered to involve the progressive accumulation of multiple genetic abnormalities, although little is known about the molecular mechanism behind the development of OSCC. Recently, amplifications and mutations of PIK3CA were reported in HNSCC.( 11 , 12 , 13 , 14 , 15 ) However, no large‐scale analysis of PIK3CA genetic alterations and their clinicopathological significance has ever been performed in OSCC.
In the present study, therefore, we analyzed 14 human OSCC cell lines as well as 108 primary OSCC tumors with regard to the frequency of mutations within hotspot regions and the changes in the genomic copy‐number of PIK3CA. Then, we evaluated if there were any statistically significant associations between PIK3CA gene status and the clinical characteristics of OSCC patients. Furthermore, to investigate which genetic event, the mutation or amplification of PIK3CA, is essential in activation of the PI3K–AKT signaling pathway, we compared the expression of PIK3CA and the phosphorylation of AKT in OSCC cell lines in which the PIK3CA and PTEN gene status was confirmed.
Materials and Methods
Cell lines. The human OSCC cell lines Ca9‐22, HSC‐2, HSC‐3 and HSC‐4 were obtained from Japan Health Science Foundation (Osaka, Japan). HOC313, HOC815, HSC‐5, HSC‐6, HSC‐7, NA, OM1, OM2, TSU and ZA, human OSCC cell lines, were generously donated by Dr T. Amagasa (Tokyo Medical and Dental University, Japan). RT7, immortalized human oral keratinocytes, and KYSE70 and TE10, human esophageal squamous cell carcinoma (ESCC) cell lines, were kindly provided by Dr N. Kamata (Hiroshima University Faculty of Dentistry, Japan) and Dr Y. Shimada (Kyoto University Graduate School of Medicine, Japan), respectively. OSCC and ESCC cell lines were maintained in Dulbecco's modified Eagle's medium (DMEM) and RPMI‐1640 medium, respectively, supplemented with streptomycin (100 µg/mL), penicillin (100 units/mL), 2 mM glutamine, and 10% fetal bovine serum (FBS). RT7 was maintained in KGM‐2 Bullet Kit (Cambrex, Walkersville, MD, USA).
To determine the growth rate of each cultured cell line, 1.0 × 104 cells/well were inoculated in 24‐well plates and the cell numbers in triplicate wells were evaluated after 7 days treatment in 0.5% FBS with LY294002 (Cell Signaling Technology, Beverly, MA, USA) by the MTT [3‐(4,5‐dimethylthiazol‐2‐yl)‐2,5‐diphenyltetrazolium bromide; DMSO] method using a Cell Counting Kit‐8 (Dojindo, Kumamoto, Japan). Results were normalized to the cell numbers in control cultures treated with DMSO alone.
Tumor samples. With the approval of local institutional review boards, 50 formalin‐fixed paraffin‐embedded OSCC samples in Japan and 58 frozen primary OSCC samples in Thailand were selected for study. Clinical and laboratory data on all of the 108 OSCC patients were collected from the patient records. The TNM classification of Union International Contre le Cancer (UICC) was used. Genomic DNA was extracted from the cell lines and the frozen tissue using a Genomic DNA Purification kit (Gentra, Minneapolis, MN) according to the manufacturer's instructions. DEXPAT (TaKaRa BIO, Otsu, Japan) was also used in the extraction of genomic DNA from the formalin‐fixed paraffin‐embedded samples. Smoking status was categorized as follows: never‐smokers, individuals who had smoked fewer than 100 cigarettes in their lifetime; former smokers, those who had quit at least 1 year before interview; current smokers, those who were currently smoking or had stopped smoking within the previous 1 year; and ever‐smokers, current and former smokers combined.
Screening for mutations. PIK3CA, epidermal growth factor receptor (EGFR) and PTEN were amplified by polymerase chain reaction (PCR) from genomic DNA (10 ng per sample) with specific oligonucleotide primers. PCR products were sequenced directly with the BigDye terminator method (Applied Biosystems, Foster City, CA, USA) on a capillary autosequencer (ABI Prism 3100) using the sequencing primers. The primers used for PCR and sequencing are shown in Table 1. All samples found to have a mutation were subsequently sequenced in the reverse direction to confirm the mutation using the antisense PCR primers. Then the mutation was further verified by sequencing of a second PCR product derived independently from the original template.
Table 1.
Primers used for PCR and sequencing for screening of mutations
| Gene | Sense | Antisense | Sequencing |
|---|---|---|---|
| PIK3CA | |||
| Exon 9 | 5′‐GATTGGTTCTTTCCTGTCTCTG‐3′ | 5′‐CCACAAATATCAATTTACAACCATTG‐3′ | 5′‐TTGCTTTTTCTGTAAATCATCTGTG‐3′ |
| Exon 20 | 5′‐TGGGGTAAAGGGAATCAAAAG‐3′ | 5′‐CCTATGCAATCGGTCTTTGC‐3′ | 5′‐TGACATTTGAGCAAAGACCTG‐3′ |
| EGFR | |||
| Exon 18 | 5′‐TCAGAGCCTGTGTTTCTACCAA‐3′ | 5′‐TGGTCTCACAGGACCACTGATT‐3′ | 5′‐TCCAAATGAGCTGGCAAGTG‐3′ |
| Exon 19 | 5′‐AAATAATCAGTGTGATTCGTGGAG‐3′ | 5′‐GAGGCCAGTGCTGTCTCTAAGG‐3′ | 5′‐GTGCATCGCTGGTAACATCC‐3′ |
| Exon 21 | 5′‐GCAGCGGGTTACATCTTCTTTC‐3′ | 5′‐CAGCTCTGGCTCACACTACCAG‐3′ | 5′‐GCTCAGAGCCTGGCATGAA‐3′ |
| PTEN | |||
| Exon 1 | 5′‐GCAGCTTCTGCCACTTCTCT‐3′ | 5′‐CATCCGTCTACTCCCACGTT‐3′ | |
| Exon 2 | 5′‐CTCCAGCTATAGTGGGGAAAA‐3′ | 5′‐CTGTATCCCCCTGAAGTCCA‐3′ | |
| Exon 3 | 5′‐TGGTGGCTTTTTGTTTGTTT‐3′ | 5′‐CATGAATCTGTGCCAACAATG‐3′ | |
| Exon 4 | 5′‐AAAGATTCAGGCAATGTTTGTT‐3′ | 5′‐TCTCACTCGATAATCTGGATGAC‐3′ | |
| Exon 5 | 5′‐GGAATCCAGTGTTTCTTTTAAATACC‐3′ | 5′‐TCCAGGAAGAGGAAAGGAAAA‐3′ | |
| Exon 6 | 5′‐ATGGCTACGACCCAGTTACC‐3′ | 5′‐TTGGCTTCTTTAGCCCAATG‐3′ | |
| Exon 7 | 5′‐TGCTTGAGATCAAGATTGCAG‐3′ | 5′‐GCCATAAGGCCTTTTCCTTC‐3′ | |
| Exon 8–1 | 5′‐GTGCAGATAATGACAAGGAATA‐3′ | 5′‐ACACATCACATACATACAAGTC‐3′ | |
| Exon 8–2 | 5′‐TTAAATATGTCATTTCATTTCTTTTTC‐3′ | 5′‐CTTTGTCTTTATTTGCTTTGT‐3′ | |
| Exon 9 | 5′‐TGTTCATCTGCAAAATGGAAT‐3′ | 5′‐CAAGTGTCAAAACCCTGTGG‐3′ | |
Screening for gene amplification. Gene amplification of PIK3CA was assessed by SYBR Green quantitative PCR. In this analysis, we used RT7 as the normal counterpart of OSCC cell lines or six normal genomic DNAs extracted from normal lymphocytes as a control for primary tumors, and KYSE70 and TE10 as a positive control. COL7A1, mapped to 3q21, was selected as a control for single copy genes. PCR was carried out with the SYBR Green PCR Master Mix (Applied Biosystems) on an ABI Prism 7900 Sequence Detection System. The primer pairs used were as follows: for PIK3CA, sense 5′‐ATCTTTTCTCAATGATGCTTGGCT‐3′, and antisense 5′‐CTAGGGTGTTTCGAATGTATG‐3′ for COL7A1, sense 5′‐ACCCAGTACCGCATCATTGTG‐3′, and antisense 5′‐TCAGGCTGGAACTTCAGTGTG‐3′.
We analyzed all results on a standard curve derived from a known concentration of sample. Six normal samples were also used in each assay, and the mean value was used to normalize the data and correct for interassay variation. The PIK3CA gene copy‐number was calculated by dividing its value by the COL7A1 values. A value exceeding fourfold the standard deviation was considered to represent high‐level DNA amplification.
Real‐time RT‐PCR. Real‐time reverse transcription‐PCR (RT‐PCR) was performed using an ABI Prism 7900 Sequence Detection System (Applied Biosystems), the SYBR Green PCR Master Mix (Applied Biosystems), and random‐primed cDNAs. The primer pairs used were as follows: for PIK3CA, sense 5′‐TTAGCTATTCCCACGCAGGA‐3′, and antisense 5′‐CACAATAGTGTCTGTGACTC‐3′ for GAPDH, sense 5′‐CGGAGTCAACGGATTTGGTCGTAT‐3′, and antisense 5′‐AGCCTTCTCCATGGTGGTGAAGAC‐3′. Expression levels of the PIK3CA gene were based on the amount of the target message relative to that of the GAPDH transcript as a control, to normalize the initial input of total RNA.
Western blot analysis. Cells were washed with phosphate‐buffered saline and lyzed in cell lysis buffer (50 mM Tris‐HCl pH 7.5, 0.5% NP‐40, 150 mM NaCl, 2 mM NaVO3, 100 mM NaF, 10 mM pyrophoric acid, and 1 mM EDTA) then boiled for 3 min. These cell lysates were analyzed by western blotting using anti‐AKT, anti‐p‐AKT (Ser‐473), anti‐p‐AKT (Thr‐308), and anti‐PTEN rabbit polyclonal antibodies (Cell Signaling Technology), and anti‐β‐actin monoclonal antibody (Sigma, St Louis, MO, USA).
Statistical analysis. The associations between PIK3CA mutations and clinicopathological characteristics were evaluated with Fisher's exact test and the chi‐squared test. A P‐value less than 0.05 was defined as being statistically significant.
Results
Identification of somatic mutations of PIK3CA in OSCC. We first investigated the genomic DNA of 14 human OSCC cell lines by direct sequencing of PCR products, and found that HSC‐2, ‐3, and ‐4 each harbored a different missense mutation in the PIK3CA gene, A3140G in exon 20, and A1634G and G1633A in exon 9, corresponding to an amino acid change of H1047R, E545G, and E545K, respectively (Table 2 and Fig. 1A). Using the same method, we also confirmed the absence of PIK3CA mutations in two ESCC cell lines, KYSE70 and TE10, used as a positive control in the subsequent real‐time PCR analysis and western blot analysis.
Table 2.
Summary of PIK3CA mutations in oral squamous cell carcinoma
| Exon | Nucleotide | Amino acid | Cell lines | Primary cases |
|---|---|---|---|---|
| 9 | G1624A | E542K* | 0 | 2 |
| 9 | G1633A | E545K* | 1 | 1 |
| 9 | A1634G | E545G | 1 | 0 |
| 9 | A1637T | Q546L | 0 | 1 |
| 20 | A3127G | M1043V | 0 | 1 |
| 20 | A3140G | H1047R* | 1 | 1 |
| 20 | G3145A | G1049S | 0 | 1 |
| 20 | G3145C | G1049R | 0 | 1 |
| Samples with mutations | 3/14 (21.4%) | 8/108 (7.4%) |
A hot spot mutation previously reported to elevate kinase activity compared with the wild type.
Figure 1.
Sequence chromatograms of missense mutations as determined by automated sequence analysis. (A) Eight different missense mutations in exons 9 and 20 of PIK3CA. (B) One missense mutation (C973G) in exon 8 of PTEN. These were directly sequenced in two repeat examinations with independent genomic PCR.
In order to compare with frequency of mutation of PIK3CA in OSCC cell lines, we screened for mutations in exons 18, 19 and 21 of the EGFR gene and exons 1–9 of the PTEN gene. No mutation or homozygous deletion of EGFR was detected in any of these cell lines. However, a missense mutation (C973G) in exon 8 of PTEN, corresponding to L325V, was found in HOC815 cells (7.1%, Fig. 1B). This mutation of the PTEN gene has not been reported previously.
Then, as the frequency of PIK3CA mutations in OSCC cell lines (21.4%) was not lower than in a previous study of various other cancers, we examined 108 OSCC primary tumors for the presence of somatic mutations in exons 9 and 20 of PIK3CA. The clinical characteristics and genetic alterations to PIK3CA in OSCC patients are summarized in Table 3. In eight specimens, we found missense mutations (7.4%): G1624A, G1633A, and A1637T in exon 9, and A3127G, A3140G, G3145A, and G3145C in exon 20, corresponding to the changes E542K, E545K, Q546L, M1043V, H1047R, G1049S, and G1049R at the amino acid level, respectively (Table 2 and Fig. 1A).
Table 3.
Clinical characteristics and PIK3CA genetic alterations in patients with oral squamous cell carcinoma
| Patient no. | Age (years) | Sex | Ethnicity | Smoking status | Betel chewing | Tumor location | Histological grade | Stage | Mutation in exon 9 and 20 | Copy number |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 75 | M | Japanese | Former smoker | – | Buccal mucosa | Well | I | G1624A | Normal |
| 2 | 58 | F | Thai | Never‐smoker | Yes | Gingiva | Moderate | I | Wild type | Normal |
| 3 | 75 | M | Thai | Never‐smoker | No | Lip | Well | I | Wild type | Normal |
| 4 | 70 | F | Thai | Never‐smoker | Yes | Tongue | Well | I | C3075T | Increased |
| 5 | 34 | F | Japanese | Current smoker | – | Tongue | Well | I | Wild type | Normal |
| 6 | 36 | F | Japanese | Never‐smoker | – | Tongue | – | I | Wild type | Normal |
| 7 | 38 | M | Japanese | Current smoker | – | Tongue | Moderate | I | Wild type | Increased |
| 8 | 42 | F | Japanese | – | – | Tongue | – | I | Wild type | Increased |
| 9 | 48 | M | Japanese | – | – | Tongue | Well | I | Wild type | Normal |
| 10 | 50 | M | Japanese | Current smoker | – | Tongue | Moderate | I | Wild type | Normal |
| 11 | 51 | F | Japanese | Current smoker | – | Tongue | – | I | Wild type | Normal |
| 12 | 53 | M | Japanese | Never‐smoker | – | Tongue | Moderate | I | Wild type | Normal |
| 13 | 54 | M | Japanese | Never‐smoker | – | Tongue | – | I | Wild type | Normal |
| 14 | 57 | F | Japanese | Never‐smoker | – | Tongue | Well | I | Wild type | Normal |
| 15 | 59 | M | Japanese | Current smoker | – | Tongue | – | I | Wild type | Normal |
| 16 | 61 | M | Japanese | Never‐smoker | – | Tongue | Well | I | Wild type | Increased |
| 17 | 61 | M | Japanese | Current smoker | – | Tongue | Moderate | I | Wild type | Normal |
| 18 | 64 | M | Japanese | Current smoker | – | Tongue | Well | I | Wild type | Normal |
| 19 | 65 | F | Japanese | Never‐smoker | – | Tongue | – | I | Wild type | Normal |
| 20 | 65 | M | Japanese | Current smoker | – | Tongue | – | I | Wild type | Normal |
| 21 | 66 | M | Japanese | Former smoker | – | Tongue | Moderate | I | Wild type | Normal |
| 22 | 69 | F | Japanese | – | – | Tongue | – | I | Wild type | Normal |
| 23 | 69 | M | Japanese | Never‐smoker | – | Tongue | Moderate | I | Wild type | Normal |
| 24 | 70 | F | Japanese | – | – | Tongue | Well | I | Wild type | Normal |
| 25 | 72 | M | Japanese | Current smoker | – | Tongue | Moderate | I | Wild type | Normal |
| 26 | 74 | M | Japanese | – | – | Tongue | Well | I | Wild type | Normal |
| 27 | 80 | M | Japanese | – | – | Tongue | – | I | Wild type | Normal |
| 28 | 80 | M | Japanese | – | – | Tongue | – | I | Wild type | Normal |
| 29 | 57 | F | Thai | Never‐smoker | No | Buccal mucosa | – | II | Wild type | Normal |
| 30 | 81 | F | Thai | Never‐smoker | Yes | Buccal mucosa | Moderate | II | Wild type | Normal |
| 31 | 85 | F | Thai | Never‐smoker | Yes | Buccal mucosa | Well | II | Wild type | Normal |
| 32 | 52 | F | Thai | Current smoker | – | Floor of mouth | – | II | Wild type | Normal |
| 33 | 55 | M | Thai | – | No | Floor of mouth | Moderate | II | A3127G | Increased |
| 34 | 76 | F | Thai | Never‐smoker | No | Floor of mouth | – | II | Wild type | Normal |
| 35 | 58 | M | Thai | Current smoker | – | Gingiva | Well | II | Wild type | Normal |
| 36 | 64 | M | Thai | – | – | Gingiva | Well | II | Wild type | Increased |
| 37 | 83 | F | Thai | Never‐smoker | Yes | Gingiva | Well | II | Wild type | Normal |
| 38 | 68 | F | Thai | Never‐smoker | Yes | Lip | – | II | Wild type | Normal |
| 39 | 79 | F | Thai | Never‐smoker | Yes | Lip | Moderate | II | Wild type | Normal |
| 40 | 59 | M | Thai | Never‐smoker | No | Retromolar | Well | II | C3075T | Normal |
| 41 | 28 | M | Thai | Current smoker | No | Tongue | Well | II | Wild type | Normal |
| 42 | 63 | M | Thai | – | – | Tongue | Moderate | II | Wild type | Increased |
| 43 | 71 | M | Thai | Current smoker | No | Tongue | Well | II | C3075T | Normal |
| 44 | 73 | F | Thai | – | Yes | Tongue | Well | II | Wild type | Normal |
| 45 | 74 | M | Thai | Current smoker | – | Tongue | Well | II | Wild type | Increased |
| 46 | – | – | Japanese | – | – | Tongue | – | II | Wild type | Normal |
| 47 | – | – | Japanese | Current smoker | – | Tongue | – | II | Wild type | Normal |
| 48 | 30 | M | Japanese | Current smoker | – | Tongue | – | II | Wild type | Increased |
| 49 | 41 | M | Japanese | Current smoker | – | Tongue | – | II | Wild type | Increased |
| 50 | 48 | M | Japanese | Current smoker | – | Tongue | – | II | Wild type | Increased |
| 51 | 49 | M | Japanese | Never‐smoker | – | Tongue | – | II | Wild type | Normal |
| 52 | 50 | M | Japanese | Current smoker | – | Tongue | Well | II | Wild type | Normal |
| 53 | 50 | F | Japanese | Current smoker | – | Tongue | Well | II | Wild type | Normal |
| 54 | 56 | M | Japanese | Never‐smoker | – | Tongue | Well | II | Wild type | Normal |
| 55 | 59 | M | Japanese | – | – | Tongue | Well | II | Wild type | Normal |
| 56 | 60 | M | Japanese | Former smoker | – | Tongue | – | II | Wild type | Normal |
| 57 | 61 | M | Japanese | – | – | Tongue | – | II | G3145A | Normal |
| 58 | 63 | M | Japanese | Current smoker | – | Tongue | – | II | Wild type | Normal |
| 59 | 63 | M | Japanese | Former smoker | – | Tongue | – | II | Wild type | Normal |
| 60 | 64 | M | Japanese | – | – | Tongue | Well | II | Wild type | Normal |
| 61 | 65 | F | Japanese | Never‐smoker | – | Tongue | – | II | Wild type | Normal |
| 62 | 66 | M | Japanese | Current smoker | – | Tongue | Moderate | II | Wild type | Normal |
| 63 | 66 | M | Japanese | – | – | Tongue | Well | II | Wild type | Normal |
| 64 | 67 | F | Japanese | Never‐smoker | – | Tongue | – | II | Wild type | Normal |
| 65 | 67 | M | Japanese | Former smoker | – | Tongue | Moderate | II | Wild type | Normal |
| 66 | 69 | M | Japanese | Current smoker | – | Tongue | Well | II | Wild type | Normal |
| 67 | 70 | F | Japanese | Current smoker | – | Tongue | Well | II | Wild type | Normal |
| 68 | 80 | M | Japanese | Former smoker | – | Tongue | – | II | Wild type | Normal |
| 69 | 51 | M | Thai | – | – | Buccal mucosa | – | III | Wild type | Increased |
| 70 | 75 | F | Thai | – | Yes | Buccal mucosa | Well | III | Wild type | Normal |
| 71 | – | M | Thai | – | – | Floor of mouth | – | III | Wild type | Increased |
| 72 | 79 | F | Thai | Never‐smoker | Yes | Gingiva | Well | III | Wild type | Normal |
| 73 | 82 | M | Thai | – | – | Gingiva | – | III | Wild type | Normal |
| 74 | 72 | M | Thai | Current smoker | – | Lip | – | III | Wild type | Normal |
| 75 | 90 | F | Thai | – | Yes | Tongue | Well | III | Wild type | Normal |
| 76 | – | – | Japanese | Current smoker | – | Tongue | Moderate | III | Wild type | Normal |
| 77 | 44 | F | Japanese | Never‐smoker | – | Tongue | Well | III | Wild type | Normal |
| 78 | 54 | F | Thai | – | – | Buccal mucosa | – | IV | Wild type | Normal |
| 79 | 59 | F | Thai | – | Yes | Buccal mucosa | Well | IV | Wild type | Normal |
| 80 | 64 | M | Thai | Current smoker | Yes | Buccal mucosa | Well | IV | Wild type | Normal |
| 81 | 65 | M | Thai | Never‐smoker | No | Buccal mucosa | Well | IV | Wild type | Normal |
| 82 | 66 | M | Thai | Current smoker | No | Buccal mucosa | Poor | IV | Wild type | Increased |
| 83 | 67 | F | Thai | – | – | Buccal mucosa | Well | IV | Wild type | Normal |
| 84 | 69 | M | Thai | Current smoker | – | Buccal mucosa | Well | IV | A1637T | Normal |
| 85 | 74 | F | Thai | Never‐smoker | Yes | Buccal mucosa | Poor | IV | Wild type | Normal |
| 86 | 75 | M | Thai | Current smoker | – | Buccal mucosa | Well | IV | A3140G | Normal |
| 87 | 77 | F | Thai | Never‐smoker | Yes | Buccal mucosa | Well | IV | G3145C | Normal |
| 88 | 78 | F | Thai | – | – | Buccal mucosa | Well | IV | Wild type | Normal |
| 89 | 81 | F | Thai | Never‐smoker | Yes | Buccal mucosa | – | IV | Wild type | Normal |
| 90 | 55 | F | Thai | – | Yes | Floor of mouth | Well | IV | Wild type | Normal |
| 91 | 55 | M | Thai | Current smoker | – | Floor of mouth | Moderate | IV | Wild type | Normal |
| 92 | 70 | M | Thai | Never‐smoker | No | Floor of mouth | Well | IV | Wild type | Normal |
| 93 | 75 | M | Thai | – | – | Floor of mouth | Poor | IV | Wild type | Normal |
| 94 | 49 | F | Thai | – | – | Gingiva | – | IV | Wild type | Normal |
| 95 | 54 | F | Thai | Never‐smoker | No | Gingiva | Well | IV | Wild type | Normal |
| 96 | 60 | M | Thai | Current smoker | No | Gingiva | Well | IV | Wild type | Normal |
| 97 | 61 | F | Thai | Never‐smoker | No | Gingiva | Well | IV | Wild type | Normal |
| 98 | 66 | M | Thai | Current smoker | No | Gingiva | Well | IV | Wild type | Increased |
| 99 | 71 | F | Thai | Current smoker | Yes | Gingiva | Moderate | IV | Wild type | Normal |
| 100 | 79 | F | Thai | Never‐smoker | Yes | Gingiva | Well | IV | G1624A | Increased |
| 101 | 82 | F | Thai | – | Yes | Gingiva | Well | IV | C3075T | Normal |
| 102 | 83 | F | Thai | Never‐smoker | Yes | Gingiva | Well | IV | Wild type | Increased |
| 103 | – | M | Thai | Current smoker | – | Gingiva | – | IV | G1633A | Increased |
| 104 | 47 | F | Thai | Current smoker | No | Hard palate | Well | IV | Wild type | Normal |
| 105 | 40 | F | Thai | Never‐smoker | Yes | Tongue | Well | IV | Wild type | Normal |
| 106 | 56 | M | Thai | Current smoker | No | Tongue | Moderate | IV | Wild type | Normal |
| 107 | 66 | M | Thai | Current smoker | – | Tongue | Poor | IV | Wild type | Normal |
| 108 | 72 | F | Thai | Never‐smoker | Yes | Tongue | Moderate | IV | Wild type | Normal |
| No. with genetic alterations (%) | 8 (7.4%) | 18 (16.7%) |
These analyses of OSCC cell lines and primary tumors led us to identify eight independent missense mutations of PIK3CA: two of three mutations in cell lines and four of eight mutations in primary tumors were hotspots with confirmed oncogenic transforming activity in vitro.( 6 , 7 , 8 , 9 ) All of the mutations detected in this study were heterozygous. However, while seven of these missense mutations have already been reported on various types of cancers, there is no information available about A1637T (Q546L). As A1637T in exon 9 of PIK3CA and C973G in exon 8 of PTEN are not stored in the COSMIC (Catalog Of Somatic Mutations In Cancer), a database of the Sanger Institute, or the NCBI SNP database (gene ID: 5290 and 5728, respectively), this is the first report of these missense mutations. Four single nucleotide polymorphisms C3075T (T1025T) in exon 20 of PIK3CA registered in the NCBI SNP database were also observed heterozygously in primary tumors.
Identification of PIK3CA amplification in OSCC. To determine the relative copy‐number of the PIK3CA gene, we employed real‐time quantitative PCR to analyze 14 OSCC cell lines and 108 OSCC primary tumors. In each analysis, we used immortalized human oral keratinocytes RT7 as the normal counterpart of OSCC cell lines and two ESCC cell lines, which were previously described to harbor a 3q26.3 gain,( 16 , 17 , 18 , 19 ) as a positive control. As shown in Fig. 2A, an increase in PIK3CA copy‐number was detected in eight OSCC cell lines (57.1%, 1.2‐ to 2.6‐fold amplification). In OSCC primary tumors, 18 specimens (16.7%, 1.3‐ to 3.4‐fold amplification) showed an increase in copy‐number (Table 3). However, the levels of PIK3CA amplification in these OSCC cell lines and primary tumors were much lower than those in the two ESCC cell lines (7.8‐ and 2.9‐fold amplification, respectively).
Figure 2.
Comparison of the effects of PIK3CA genetic alterations and LY294002 on the PI3K–AKT signaling pathway in a panel of OSCC cell lines. (A) Comparison of copy‐number (left) and expression (right) of PIK3CA by real‐time quantitative PCR. These results were normalized to the copy‐number or expression levels of the PIK3CA gene in RT7 as a normal counterpart. (B) Comparison of AKT activation and PTEN protein expression levels by western blot analysis. To determine the increased phosphorylation of AKT, protein levels of total AKT in same samples were evaluated. Cell lines were cultured in 10% serum, and lysates prepared from them were immunoblotted with the indicated antibodies. (C) The effect of LY294002 on growth in vitro. Cell lines were cultured in 0.5% serum at the indicated concentration of LY294002 diluted in DMSO, and evaluated by the MTT method 7 days after treatments. These results were normalized to the cell numbers in control cultures treated with DMSO alone. (D) The effect of LY294002 on AKT activation. Cell lines were treated with 20 µM LY294002 or DMSO alone in medium containing 0.5% serum for 2 h, and evaluated by western blot analysis using the antibodies indicated.
Relationship between PIK3CA genetic alterations and clinical characteristics. We analyzed the correlation of PIK3CA mutations in exons 9 and 20 with clinical data of OSCC patients, and found a statistically significant association between the frequency of mutation and stage of disease (P = 0.042, Fisher's exact test) (Table 4). The highest frequency of PIK3CA mutations was observed in stage IV of the disease (16.1%, 5/31), although all of the patients with stage IV disease were Thais (Table 3). There was, however, no significant difference between Japanese and Thais with regard to the frequency of PIK3CA mutations in a subset of patients with early stage (stage I–III) disease: 4% (2/50) versus 2% (1/27), respectively. In contrast, amplification of PIK3CA was detected in 14.3% (4/28), 17.5% (7/40), 22.2% (2/9), and 16.1% (5/31) of patients with stage I, II, III, and IV disease, respectively (Table 3). These frequencies were markedly lower than those reported for HNSCC,( 11 , 12 , 13 ) and showed no association with disease stage (P = 0.952, chi‐squared test). Therefore, our findings suggest that mutations in PIK3CA may be an oncogenic event at the advanced stages of OSCC, and that amplification of the gene may precede mutations in the development of OSCC.
Table 4.
Correlations between PIK3CA mutations and clinicopathological characteristics of patients with oral squamous cell carcinoma
| Characteristic | PIK3CA mutation | No PIK3CA mutation | P‐value |
|---|---|---|---|
| Total number (%) | 8 (7.4%) | 100 (92.6%) | |
| Age (years) [median (range)] | 75 (55–79) | 64 (28–90) | |
| Sex | |||
| Male | 6 | 54 | P = 0.462* |
| Female | 2 | 43 | |
| Unknown | 0 | 3 | |
| Ethnicity | |||
| Japanese | 2 | 48 | P = 0.282* |
| Thai | 6 | 52 | |
| Smoking status | |||
| Ever‐smokers | 4 | 41 | P = 0.691* |
| Never‐smokers | 2 | 33 | |
| Unknown | 2 | 26 | |
| Location | |||
| Buccal mucosa | 4 | 14 | P = 0.123 † |
| Floor of mouth | 1 | 7 | |
| Gingiva | 2 | 14 | |
| Hard palate | 0 | 1 | |
| Lip | 0 | 4 | |
| Retromolar | 0 | 1 | |
| Tongue | 1 | 59 | |
| Histological grading | |||
| Well | 5 | 45 | P = 0.674 † |
| Moderate | 1 | 18 | |
| Poor | 0 | 4 | |
| Unknown | 2 | 33 | |
| Disease stage | |||
| I–III | 3 | 74 | P = 0.042* |
| IV | 5 | 26 | |
| PIK3CA amplification | |||
| Positive | 3 | 15 | P = 0.127* |
| Negative | 5 | 85 | |
Fisher's exact test;
† chi‐squared test.
There was no relationship between the presence of PIK3CA mutations and sex, ethnicity, smoking behavior, location, histological grading, or PIK3CA amplification (Table 4). In addition, there was no association between betel chewing and PIK3CA mutations in Thai patients (P ≥ 0.999, Fisher's exact test).
We could not analyze the prognostic significance because the complete survival data were not included in our clinical data.
Effects of PIK3CA genetic alterations on PIK3CA expression and AKT activation in OSCC cell lines. To investigate which genetic event, mutation or amplification of PIK3CA, is crucial to the activation of the PI3K–AKT signaling pathway in OSCC, we analyzed the expression of PIK3CA and the phosphorylation of AKT in a panel of cell lines in which the PIK3CA and PTEN gene status was conformed. In the real‐time RT‐PCR analysis, we noted a weak tendency for high levels of PIK3CA mRNA in OSCC cell lines with PIK3CA amplification (Fig. 2A): five of eight OSCC cell lines with increased PIK3CA copy‐numbers expressed the PIK3CA transcript at more than twofold the level found in RT7, while the expression level of PIK3CA in all of six OSCC cell lines with a normal PIK3CA copy‐number changed less than twofold. However, the level of PIK3CA expression in the two ESCC cell lines was much higher than that in the OSCC cell lines.
Western blot analysis demonstrated that the phosphorylation of AKT at residues Ser‐473 and Thr‐308 was markedly increased in HSC‐2, ‐3, and ‐4 with PIK3CA mutations (Fig. 2B). Ca9‐22, HOC815, and ZA, which were high PIK3CA expressers with amplification in OSCC cell lines, also showed phosphorylated AKT. In addition, KYSE70 and TE10 were high expressers with PIK3CA amplification, whereas only a very weak phosphorylation of AKT was detected in both ESCC cell lines. The PTEN protein was sufficiently expressed in 13/14 of OSCC cell lines and RT7, but not in HOC815 harboring a PTEN mutation. Notably, no clear correlation was found between protein expression levels of PTEN and the activation of AKT in these cell lines. Phosphorylation of AKT detected in RT7 was thought to have been increased by recombinant EGF added in the culture medium. Although NA without a genetic alteration of PIK3CA or PTEN also showed phosphorylated AKT, it was unknown why the phosphorylation increased in this cell line. Hence, these results indicate that the mutation, rather than an increase in the copy‐number, of PIK3CA may play a crucial role in the activation of the PI3K–AKT signaling pathway.
Effects of LY294002 on cell growth and AKT activation in OSCC cell lines. Next, we compared the effects of LY294002, a commonly used PI3K inhibitor, on RT7 and OSCC cell lines with and without mutant forms of PIK3CA in vitro. LY294002 was shown to inhibit cell growth in a dose‐dependent manner. However, no difference in the inhibitory effects of LY294002 on in vitro growth rates was seen among cell lines analyzed, regardless of the mutation, expression, or amplification of PIK3CA (Fig. 2C). LY294002 inhibited the phosphorylation of AKT in all cell lines analyzed (Fig. 2D). Thus, the inhibitory effects of LY294002 on cell growth were not related to the existence of PIK3CA genetic abnormalities.
Discussion
The PI3K–AKT pathway, as well as Ras and p53, are important signaling pathways contributing to cancer development, and are aberrantly activated in many types of cancers by gain‐ or loss‐of‐function of components of this pathway, such as PTEN, AKT and PIK3CA.( 1 , 2 , 3 ) Here we performed a large‐scale mutational analysis of cell lines and primary tumors of OSCC, and found a significant correlation between the advanced stage of OSCC and the frequency with which PIK3CA is mutated in exons 9 and 20 (P = 0.042). The present study is the first report to mention a significant correlation between PIK3CA mutations and disease stage in OSCC. The frequency of mutations of the PIK3CA gene has been reported to be 32% in colon cancer, 3–27% in brain tumors, 4–25% in gastric cancer, 8–40% in breast cancer, 4% in lung cancer, 4–7% in ovarian cancer, and 11% in HNSCC.( 4 , 15 , 20 , 21 , 22 , 23 ) In our study, the frequency of mutations in cell lines (21.4%) and primary tumors (7.4%) of OSCC was not as high as in previous studies, although PIK3CA was mutated at a relatively high frequency in stage IV (16.1%), suggesting that PIK3CA mutations may be a late event in genomic aberration involved in the progression of OSCC. The frequency of PIK3CA mutations in OSCC specimens could be even lower than that observed in OSCC cell lines, as activation of the PI3K–AKT pathway may have conferred a selective growth advantage in vitro,( 6 , 7 , 8 , 9 ) leading to the successful establishment of cell lines. However, it has also been reported that PIK3CA mutations were detected in the early lesions of various cancers, such as intraductal carcinoma of breast cancer (26.9%), dysplastic nodule of hepatocellular carcinoma (35.6%), and early gastric cancer (6.5%).( 24 ) Taken together, those results suggest that the clinicopathological significance of PIK3CA mutations may differ according to the type of cancer.
Amplification of the PIK3CA gene has been reported in various human cancers,( 1 , 25 ) and described as a genomic aberration in the early stages of HNSCC.( 11 , 12 ) Consistent with these reports, our study showed that the copy‐number of PIK3CA was also increased in the early stages of OSCC, independent of the stage of the disease (P = 0.952). In addition, the phosphorylation of AKT was considerably increased in OSCC cell lines with PIK3CA mutations compared with those without mutations, despite the amplification. Recently, Samuels et al. clearly showed that overexpression of a hotspot mutant PIK3CA remarkably increased the phosphorylation of AKT in comparison to overexpression of wild‐type PIK3CA in vitro.( 9 ) In their study, the increased AKT activation was due to increased PIK3CA kinase activity in the mutant form, not increased protein levels of PIK3CA, indirectly supporting our findings in vitro. Another point, which should be mentioned, was that the levels of increase in the copy‐number of PIK3CA were very low in cell lines and primary tumors of OSCC as compared with ESCC cell lines.( 16 , 17 , 18 , 19 ) Therefore, the activation of PIK3CA due to mutation, rather than amplification, may strongly contribute to the dysregulation of the PI3K–AKT signaling pathway in the advanced progression, especially stage IV, of OSCC.
One of the most impressive recent developments in cancer therapy has been the use of kinase inhibitors such as imanitib (Gleevec), trastzumab (Herceptin), and gefitinib (Iressa).( 26 , 27 , 28 ) Sensitivity to gefitinib, an EGFR‐specific tyrosine kinase inhibitor, was reported to be significantly higher in non‐small‐cell lung cancer patients with EGFR mutations than those without these mutations.( 29 , 30 ) A mutation in the kinase domain of EGFR was reported to be rare in HNSCC (1–7%),( 31 , 32 ) and was not detected in OSCC cell lines examined in this study. Thus, we look forward to development of molecules, other than EGFR, as therapeutic targets in OSCC. PIK3CA could be a potential target for cancer therapy, although its regulated signaling pathway is very important even in normal cells. LY294002, a commonly used pharmacological inhibitor of the cAMP‐independent PI3K–AKT signal transduction pathway, has been shown to inhibit the growth of tumors with an activated PI3K–AKT pathway in vitro and in vivo.( 9 , 33 ) However, our in vitro study demonstrated that LY294002 inhibited cell growth and AKT activation in tumor cells regardless of the presence of PIK3CA mutations. These in vitro data strongly suggest that the development of PIK3CA mutation‐specific inhibitors is important for cancer therapy in the clinical setting.
In summary, our findings suggest that: (i) PIK3CA mutations may be an oncogenic aberration at advanced stages of OSCC; (ii) the mutation, rather than increased DNA copy‐number, of PIK3CA is likely to function as an oncogene in OSCC; and (iii) as PIK3CA could be a promising target for cancer therapies, further development of PIK3CA mutation‐specific inhibitors is warranted because of the central importance of the PI3K–AKT nexus in normal and tumor cells.
Acknowledgments
We are grateful to Professor Yusuke Nakamura (Human Genome Center, Institute of Medical Science, The University of Tokyo) and Professor Masaki Noda (Hard Tissue Genome Research Center, Tokyo Medical and Dental University) for continuous encouragement throughout this work.
This study was supported in part by Grants‐in‐Aid for Scientific Research and Scientific Research on Priority Areas, and a 21st Century Center of Excellence (COE) Program for Molecular Destruction and Reconstitution of Tooth and Bone from the Ministry of Education, Culture, Sports, Science, and Technology, Japan; and a grant from Core Research for Evolutional Science and Technology (CREST) of Japan Science and Technology Corporation (JST).
References
- 1. Vivanco I, Sawyers CL. The phosphatidylinositol 3‐kinase AKT pathway in human cancer. Nat Rev Cancer 2002; 2: 489–501. [DOI] [PubMed] [Google Scholar]
- 2. Vogelstein B, Kinzler KW. Cancer genes and the pathways they control. Nat Med 2004; 10: 789–99. [DOI] [PubMed] [Google Scholar]
- 3. Cully M, You H, Levine AJ, Mak TW. Beyond PTEN mutations: the PI3K pathway as an integrator of multiple inputs during tumorigenesis. Nat Rev Cancer 2006; 6: 184–92. [DOI] [PubMed] [Google Scholar]
- 4. Samuels Y, Wang Z, Bardelli A et al. High frequency of mutations of the PIK3CA gene in human cancers. Science 2004; 304: 554. [DOI] [PubMed] [Google Scholar]
- 5. Bader AG, Kang S, Zhao L, Vogt PK. Oncogenic PI3K deregulates transcription and translation. Nat Rev Cancer 2005; 5: 921–9. [DOI] [PubMed] [Google Scholar]
- 6. Ikenoue T, Kanai F, Hikiba Y et al. Functional analysis of PIK3CA gene mutations in human colorectal cancer. Cancer Res 2005; 65: 4562–7. [DOI] [PubMed] [Google Scholar]
- 7. Isakoff SJ, Engelman JA, Irie HY et al. Breast cancer‐associated PIK3CA mutations are oncogenic in mammary epithelial cells. Cancer Res 2005; 65: 10992–1000. [DOI] [PubMed] [Google Scholar]
- 8. Kang S, Bader AG, Vogt PK. Phosphatidylinositol 3‐kinase mutations identified in human cancer are oncogenic. Proc Natl Acad Sci USA 2005; 102: 802–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Samuels Y, Diaz LA Jr, Schmidt‐Kittler O et al. Mutant PIK3CA promotes cell growth and invasion of human cancer cells. Cancer Cell 2005; 7: 561–73. [DOI] [PubMed] [Google Scholar]
- 10. Nomura K, Sobue T, Nakatani H et al. Number of deaths and proportional mortality rates from malignant neoplasms by site in Japan (2003). In: Editorial Board of Cancer Statistics in Japan, eds. Cancer Statistics in Japan 2005. Tokyo: National Cancer Center, 2005; 36–9. Available from: http://www.ncc.go.jp/en/statistics/index.html [Google Scholar]
- 11. Redon R, Muller D, Caulee K, Wanherdrick K, Abecassis J, Du Manoir S. A simple specific pattern of chromosomal aberrations at early stages of head and neck squamous cell carcinomas: PIK3CA but not p63 gene as a likely target of 3q26‐qter gains. Cancer Res 2001; 61: 4122–9. [PubMed] [Google Scholar]
- 12. Woenckhaus J, Steger K, Werner E et al. Genomic gain of PIK3CA and increased expression of p110alpha are associated with progression of dysplasia into invasive squamous cell carcinoma. J Pathol 2002; 198: 335–42. [DOI] [PubMed] [Google Scholar]
- 13. Pedrero JM, Carracedo DG, Pinto CM et al. Frequent genetic and biochemical alterations of the PI 3‐K/AKT/PTEN pathway in head and neck squamous cell carcinoma. Int J Cancer 2005; 114: 242–8. [DOI] [PubMed] [Google Scholar]
- 14. Lin M, Smith LT, Smiraglia DJ, Kazhiyur‐Mannar R et al. DNA copy number gains in head and neck squamous cell carcinoma. Oncogene 2006; 25: 1424–33. [DOI] [PubMed] [Google Scholar]
- 15. Qiu W, Schonleben F, Li X et al. PIK3CA mutations in head and neck squamous cell carcinoma. Clin Cancer Res 2006; 12: 1441–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Shinomiya T, Mori T, Ariyama Y et al. Comparative genomic hybridization of squamous cell carcinoma of the esophagus: the possible involvement of the DPI gene in the 13q34 amplicon. Genes Chromosomes Cancer 1999; 24: 337–44. [PubMed] [Google Scholar]
- 17. Imoto I, Pimkhaokham A, Fukuda Y et al. SNO is a probable target for gene amplification at 3q26 in squamous‐cell carcinomas of the esophagus. Biochem Biophys Res Commun 2001; 286: 559–65. [DOI] [PubMed] [Google Scholar]
- 18. Imoto I, Yuki Y, Sonoda I et al. Identification of ZASC1 encoding a Kruppel‐like zinc finger protein as a novel target for 3q26 amplification in esophageal squamous cell carcinomas. Cancer Res 2003; 63: 5691–6. [PubMed] [Google Scholar]
- 19. Sonoda I, Imoto I, Inoue J et al. Frequent silencing of low density lipoprotein receptor‐related protein 1B (LRP1B) expression by genetic and epigenetic mechanisms in esophageal squamous cell carcinoma. Cancer Res 2004; 64: 3741–7. [DOI] [PubMed] [Google Scholar]
- 20. Broderick DK, Di C, Parrett TJ et al. Mutations of PIK3CA in anaplastic oligodendrogliomas, high‐grade astrocytomas, and medulloblastomas. Cancer Res 2004; 64: 5048–50. [DOI] [PubMed] [Google Scholar]
- 21. Campbell IG, Russell SE, Choong DY et al. Mutation of the PIK3CA gene in ovarian and breast cancer. Cancer Res 2004; 64: 7678–81. [DOI] [PubMed] [Google Scholar]
- 22. Li VS, Wong CW, Chan TL et al. Mutations of PIK3CA in gastric adenocarcinoma. BMC Cancer 2005; 5: 29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Wang Y, Helland A, Holm R, Kristensen GB, Borresen‐Dale AL. PIK3CA mutations in advanced ovarian carcinomas. Hum Mutat 2005; 25: 322. [DOI] [PubMed] [Google Scholar]
- 24. Lee JW, Soung YH, Kim SY et al. PIK3CA gene is frequently mutated in breast carcinomas and hepatocellular carcinomas. Oncogene 2005; 24: 1477–80. [DOI] [PubMed] [Google Scholar]
- 25. Shayesteh L, Lu Y, Kuo WL et al. PIK3CA is implicated as an oncogene in ovarian cancer. Nat Genet 1999; 21: 99–102. [DOI] [PubMed] [Google Scholar]
- 26. Druker BJ, Talpaz M, Resta DJ et al. Efficacy and safety of a specific inhibitor of the BCR‐ABL tyrosine kinase in chronic myeloid leukemia. N Engl J Med 2001; 344: 1031–7. [DOI] [PubMed] [Google Scholar]
- 27. Couzin J. Cancer drugs. Smart weapons prove tough to design. Science 2002; 298: 522–5. [DOI] [PubMed] [Google Scholar]
- 28. Green MR. Targeting targeted therapy. N Engl J Med 2004; 350: 2191–3. [DOI] [PubMed] [Google Scholar]
- 29. Dowell JE, Minna JD. The impact of epidermal‐growth‐factor‐receptor mutations in response to lung‐cancer therapy. Nat Clin Pract Oncol 2004; 1: 2–3. [DOI] [PubMed] [Google Scholar]
- 30. Paez JG, Janne PA, Lee JC et al. EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science 2004; 304: 1497–500. [DOI] [PubMed] [Google Scholar]
- 31. Lee JW, Soung YH, Kim SY et al. Somatic mutations of EGFR gene in squamous cell carcinoma of the head and neck. Clin Cancer Res 2005; 11: 2879–82. [DOI] [PubMed] [Google Scholar]
- 32. Loeffler‐Ragg J, Witsch‐Baumgartner M, Tzankov A et al. Low incidence of mutations in EGFR kinase domain in Caucasian patients with head and neck squamous cell carcinoma. Eur J Cancer 2006; 42: 109–11. [DOI] [PubMed] [Google Scholar]
- 33. Hu L, Zaloudek C, Mills GB, Gray J, Jaffe RB. In vivo and in vitro ovarian carcinoma growth inhibition by a phosphatidylinositol 3‐kinase inhibitor (LY294002). Clin Cancer Res 2000; 6: 880–6. [PubMed] [Google Scholar]