ROLE OF MOLECULAR PATHOLOGY IN TARGETED THERAPY IN CHOLANGIOCARCINOMA

Background. Cholangiocarcinoma (CCA) is a heterogeneous group of highly aggressive malignancies of the biliary tract. CCAs are classified as intrahepatic (iCCAs) and extrahepatic (eCCAs). Till now, the prognosis is still poor and novel treatment strategies in CCA are needed. While in eCCA the main encountered alterations (KRAS, TP53, SMAD4) are still un-targetable, in iCCA IDH1 mutations are detected in 20% and FGFR2 alterations in 10-15% of cases, and both of these two genetic alterations are druggable with therapies already available in European countries. A standardized molecular profiling of CCA samples will be necessary and pathologists play an essential role in assessing the viability of samples for NGS analysis and in performing NGS or other molecular techniques (such as FISH) on the basis of the tissue sample’s quality. In the first part of this PhD project, we put a specific focus on the advantages and disadvantages of NGS and FISH methodologies in CCA, comparing the output obtained with these different technologies, which will to our mind produce the gold standard for gene fusion analysis in the routine clinical setting. The second part of PhD project focuses, for the first time, on the Claudin18.2 expression in a wide cohort of Western CCAs. Materials and Methods. We retrospectively collected a total of 249 iCCAs and 58 eCCAs, which were referred to the Unit of Medical Oncology of the Istituto Oncologico Veneto (IOV) of Padua. In the first part of this PhD project, among 249 iCCAs, the molecular profiling was performed: on one side, the identification of IDH1 single nucleotide mutations was investigated by a real-time PCR and, on the other side, the identification of FGFR2 rearrangements was performed by: i) an anchored multiplex RNA-seq NGS in all cases, and ii) a break-apart FISH. In the second part of this PhD project, Claudin 18.2 expression was assessed by immunohistochemistry in a cohort composed of 147 iCCAs and 58 eCCAs. From the immunohistochemical analysis, we have used two preliminary cut-offs of immunohistochemical positivity: the first one based on more than 50% positive neoplastic cells and the second one based on more than 75% positive neoplastic cells. Results. In 249 iCCAs, the prevalence of IDH1 mutations and FGFR2 rearrangements was 16.1% and 12.2%, respectively. Genetic alterations on the FGFR2 gene were FGFR2 rearrangements (9.2%), amplification (0.4%), and SNV (2.4%). Using a combined approach based on RNA-seq NGS and break-apart FISH, the definition of the FGFR2 status was possible in 245 out of the 249 cases (98.4%) due to a lack of material/low cellularity and/or low quality of the extracted RNA. The comparison between RNA-seq NGS and break-apart FISH analysis reveals a moderate concordance (k=0.440) and break-apart FISH has failed in 14% of cases. In the second part of this PhD project, in 147 iCCAs, Claudin18.2 expression was found in 12.9% and its expression was significantly associated with: i) histopathological large duct type iCCA (p=0.0008), and ii) macroscopic intraductal/periductal-infiltrating pattern (p=0.0001). According to the 50% and 75% preliminary cut-offs, in iCCA, 5.4% and 2% of cases were Claudin18.2 positive, respectively. Among the 58 eCCAs, the Claudin18.2-positive frequency was higher than in iCCAs (54.4% of eCCAs were found positive). Claudin18.2 expression was significantly associated with the histopathological intraductal type eCCA (p<0.0001). According to 50% and 75%cut-offs, in eCCA 22.4% and 8.6% of cases were Claudin18.2 positive, respectively. Conclusions. On one side, our results suggest that scant biopsies of CCA are well suited for RNA-seq NGS to identify rearrangements on FGFR2, but if the tumor content is not sufficient, a break-apart FISH can be performed. On the other side, Claudin18.2 positivity is higher in eCCAs than iCCAs and these results may pave the way to move beyond traditional therapies.