Prognostic and clinical impact of PIK3CA mutation in gastric cancer: pyrosequencing technology and literature review

This study retrospectively enrolled 208 gastric cancer patients who underwent resection at Kumamoto University Hospital between January 2001 and August 2010. Patients who underwent palliative resection and/or whose tissue samples were unavailable were excluded, but patients positive on peritoneal washing cytology were included. Patients were followed-up at 1- to 3-month intervals until death or 31 March 2015, whichever came first. Disease-free survival was defined as the time from the date of surgery to the date of cancer recurrence or death. Tumors were staged according to the American Joint Committee on Cancer Staging Manual (7th edition) [15]. Overall survival was defined as the time from the date of the operation to the date of death. All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1964 and later versions. Informed consent or substitute for it was obtained from all patients for being included in the study. The study procedures were approved by the Ethics Committee for Epidemiological and General Research at the Faculty of life Science, Kumamoto University (Approval number: Ethic 559). Throughout this article, the definition of “prognostic marker” is consistent with REMARK Guidelines [16].

Genomic DNA was extracted from paraffin-embedded tissue specimens of surgically resected gastric cancers. Tumors areas were marked on hematoxylin and eosin stained slides by one pathologist (Y. Baba). Genomic DNA was extracted from tumor lesions enriched with neoplastic cells, without adjacent normal tissue, using FFPE Kits (Qiagen, Duesseldorf, Germany).

PIK3CA mutations were evaluated as previously described [14]. The exon 9 PCR primers were 5′-biotin-AACAGCTCAAAGCAATTTCTACACG-3′ (forward, PIK3CA 9-F) and 5′-ACCTGTGACTCCATAGAAAATCTTT-3′ (reverse, PIK3CA9-R). The exon 20 PCR primers were 5′-biotin-CAAGAGGCTTTGGAGTATTTCA-3′ (forward, PIK3CA 20-F) and 5′-CAATCCATTTTTGTTGTCCA-3′ (reverse, PIK3CA 20-R). In the PIK3CA exon 9 pyrosequencing assays, the presence of mutations was routinely confirmed using three different sequencing primers and by the creation of a frameshifted reading of a mutant sequence relative to a wild-type sequence. The primers PIK3CA 9-RS1 (5′-CCATAGAAAATCTTTCTCCT-3′; nucleotide dispensation order, ATCGACTACACTGACTGACTGACTGACTGACTGACTG), PIK3CA 9-RS2 (5′ TTCTCCTTGCTTCAGTGATTT-3′; nucleotide dispensation order, ATACACATGTCAGTCAGACTAGCTAGCTAGCTAG) and PIK3CA 9-RS3 (5′-TAGAAAATCTTTCTCCTGCT-3′; nucleotide dispensation order, ATAGCACTGACTGACTGACTACTGACTGACTGACTG) could detectc.1634A > G, c.1624A > G, and c.1624G > A mutations, respectively. For PIK3CA exon 20, the primer PIK3CA 20-RS (5′-GTTGTCCAGCCACCA-3′; nucleotide dispensation order, CTGACGATACTGTGCATCATATGCATGCATGCATGCATGC) was used to detect the mutations c.3140A > G and c.3139C > T. Nucleotide dispensation orders were designed so that if any of the common mutations were present, it caused a shift in the reading frame and resulted in an additional peak or peaks following the mutated nucleotide. Representative pyrograms of wild-type and mutant exons 9 and 20 are shown in Fig. 1.

The results were statistically analyzed by JMP software (Version 9, SAS Institute, Cary, NC, USA). All p-values were two-sided. The means were compared by t test, assuming unequal variances. The proportional differences among the characteristics were evaluated by Pearson’s chi-squared test (χ ²). Survival time was determined by the Kaplan–Meier method and compared using the log-rank test. The hazard ratio for PIK3CA mutations on mortality was assessed by Cox regression modeling. To assess whether any of these variables was associated with the relationship between the PIK3CA mutations and prognosis, they were cross-multiplied with PIK3CA mutations and subjected to the Wald test.

Of the 208 patients who had undergone curative resection of gastric cancer, 25 (12 %) were found pyrosequencing technology to have PIK3CA exon 9 and 20 mutations (Fig. 1). Ten patients were positive for c.1634A > G (p.E545G), 10 for c.1624G > A (p.E542K), 13 for c. c.1633G > A (p.E545K), nine for c. 3139C > T (p.H1047R), and one for c. 3140A > G (p.H1047Y) mutations. Three tumors were positive for mutations at all three nucleotides in exon 9, three were positive for mutations at two nucleotides in exon 9. Fifteen samples were positive for mutations in exon 9 alone, four for mutations in exon 20 alone, and six were positive for mutations in both exons.

PIK3CA mutations were not significantly associated with any clinical, epidemiologic, or pathologic characteristic (Table 1). We previously reported that the long interspersed nucleotide element-1 (LINE-1) methylation level in gastric cancer was a surrogate marker of global DNA methylation and genome instability [17, 18]. However, LINE-1 methylation level was not associated with PIK3CA mutations (Table 1).

We assessed the influence of PIK3CA mutations on clinical outcome in patients with curatively resected gastric cancer. During follow-up of the 208 patients, 32 patients experienced gastric cancer recurrence and 69 died. The median follow-up time for censored patients was 5.0 years. Kaplan–Meier analysis showed no significant differences in disease-free survival (log rank P = 0.84) and overall survival (log rank P = 0.74) between patients with PIK3CA mutations and those with PIK3CA wild type (Fig. 2).

We also examined whether the influence of PIK3CA mutations on overall survival was modified by any clinical or pathological variable. All tested variables did not significantly interact with the relationship between PIK3CA mutations and overall survival (p for all interactions >0.05, Fig. 3). Moreover, analysis of various strata showed no significant difference in hazard ratio for overall survival between patients with PIK3CA mutations and those with PIK3CA wild type (Fig. 3).