Gall bladder is the most aggressive malignancy of gastrointestinal tract with a wide geographical variation as well as heterogeneity at clinical and molecular level. Despite several developments in understanding the molecular pathogenesis of GBC, there has been no considerable progress in targeted therapy or personalized approach in GBC management. KRAS mutation which is one of the predictive marker for anti-EGFR (epidermal growth factor receptor) therapy varies widely in GBC from 0 to 41% in different studies [9–16]. We found 5.4% KRAS mutation in GBC. In another previous report from our group in which we studied 49 cases of GBC, KRAS mutation was seen in 2% cases [11]. Javle et al and Pai et al from United States also reported KRAS mutation frequency in similar range of 5.5% and 7% respectively in GBC. [12, 13]. There had been some reports from Japan with variable of KRAS mutation frequency by Asai et al, Yokoyama et al and Noguchi et al with 2.8%, 9.0% and 14.3% respectively [14–16]. BRAF, another predictive marker as well as candidate for targeted therapy shows most common mutation in codon 600 (90% of all BRAF mutations) with substitution of valine to glutamine (V600E) being the most common mutation (~ 90%) of all codon 600 mutations [17]. In GBC, BRAF mutation varies from 0–33% [18–20]. Yokoyama et al. from Japan, Goldenberg et al and Pai et al from United States, also did not show any BRAF mutation in GBC [13, 15, 19]. In the present study as well as a previous report from our group did not show any BRAF mutation in GBC [11]. PIK3CA mutation is another candidate for targeted therapy which has been reported in lower frequency than KRAS and BRAF in GBC in the range of 0 to 21.5% [14–16, 21]. Some reports from United States have shown PIK3CA mutation between 3.5 to 14% in GBC [9, 22] while Yokoyama et al. [15] did not find any PIK3CA mutation in their study. We also did not find any PIK3CA mutation in exons 9 and 20 in this study. Thus, the above findings suggest that molecular profile of GBC similar to its incidence, varies widely in different regions of the world. In India, a country with one of the highest GBC incidence, common candidates for targeted therapy such as KRAS, BRAF and PIK3CA show a very low mutation rate of KRAS mutation rendering these patients to be suitable for anti-EGFR therapy.
KRAS mutation has been regarded as a poor prognostic marker in different cancers. In GBC there are very few reports on association of KRAS mutation and clinicopathological features. It has been found to be more frequent in higher grade and advanced stage tumors. Asai et al showed that all KRAS mutant GBCs were poorly differentiated in his study [14]. In a recent report, Bagante et al reported that KRAS mutant GBC were more likely to show perineural invasion and positive resection margins as well as lower 5 year overall survival of 13% [23]. In one of our earlier report of 49 GBC cases, KRAS mutation was seen in 2% cases having poorly differentiated morphology with pathological stage of T2 [11]. In the present study we found a significant association of KRAS mutation with age where all 6 cases of KRAS mutant GBC were > 50 years age (p value = 0.02). Fifty percent of the KRAS mutant GBCs were moderate to poorly differentiated with nearly 67% of them harboring perineural invasion, lymph node metastasis and being stage III tumors. The KRAS mutant GBC also showed lower mean survival of 68 months than KRAS wild GBC with a mean survival of 89 months. These findings suggest that present of KRAS mutation confers a worse prognosis as well as these patients loose the opportunity of being candidates for targeted therapy.
MMR deficiency is another important prognostic and predictive biomarker for certain chemotherapy and immune check-point blockers [24, 25]. Microsatellite instability (MSI) as demonstrated by molecular methods like PCR and fragment length analysis or deficient MMR expression demonstrated by IHC has a comparable sensitivity (77–100%) and specificity (98–100%). The frequency of microsatellite instable or MMR deficient GBC ranges between 0 to 42% with IHC and / or molecular methods [26–31] where Saetta et al 18] did not find any MMR deficient tumor but Yanagisawa et al [29] reported 41% MMR deficient GBC. Moy et al reported nearly 8% MMR loss in GBC but did not find any significant association between MMR deficient GBC and clinicopathological features such as age, grade, stage, mucinous histology and TILs, however they found these patients to be older with a median age of 71 years having moderate to poorly differentiated histology but with an overall lower tumor stage 28]. Conversely Nagahashi et al reported a frequency of MSI high to be 42% (overall MSI 38%) with 75% of instable tumors being well differentiated, 19% moderately differentiated and 6% poorly differentiated tumors [30]. Geoppert et al reported 1.4% instable GBC with lymphnode metastasis, perineural invasion and stage III tumor but a longer overall survival [31]. We found MMR loss in 27.6% of GBC with 65% of tumors having well differentiated histology but 58% having advanced stage disease (stage III). Ten percent of MMR deficient GBC showed KRAS mutation in comparison to 4% MMR proficient GBC (p value = 0.2). The wide variability in MMR deficient cases in GBC and its inconsistent relation with histological features may be due to differences in ethnic, geographical and dietary variations. No significant association has been found between KRAS mutation and MMR deficient GBCs. The TILs are regarded as one of the predictors for immune check-point therapy and since MMR deficient cancers have a tendency to harbour increased TILs, they are also potential candidates for immunotherapy, however we did not find any significant difference in TILs between MMR deficient and proficient tumors.