Mutation Profiling in Cholangiocarcinoma: Prognostic and Therapeutic Implications

Case Selection

Surgically resected or tumor core biopsy FFPE specimens were obtained for 75 patients with CCA. All patients selected were those with pathologically confirmed CCA and a minimum of 3 months of follow-up. Patients signed an informed consent that covered review of medical records, and studies for correlated research. The study was approved by the Institutional Review Board of MD Anderson Cancer Center. Patient demographics, clinical data, survival data and treatment history were retrieved from medical records. For these cases, if adequate material was procured, somatic mutation analysis was first performed using a NGS panel of hotspot regions in 46 cancer-related genes described below. In cases where targetable mutations were not identifiable using this technology or if patients were being screened for clinical trials that required a wider NGS panel, Foundation Medicine (Boston, MA) performed additional sequencing with the IIlumina Hiseq2000 Platform.

Tumor samples

The paraffin embedded blocks were sectioned, and hematoxylin & eosin (H&E) stained slides were reviewed by surgical pathology to confirm the tumor content in each section. Ten serial sections (4 µm) were cut from selected tissue blocks and areas with tumor tissue were macrodissected from those slides using the H&E slides as templates. Clinical data were abstracted from all the patients. Approval for the study was obtained from the institutional review board at MD Anderson Cancer Center.

DNA Extraction

The pathologic diagnosis of each case of CCA was confirmed on routine H & E slides. All samples sent for DNA extraction contained a minimum of 20% DNA derived from tumor cells. DNA was extracted from 40 mm of FFPE tissue using the Maxwell 16 FFPE Plus LEV DNA Purification kit (Promega) and quantified using a standardized PicoGreen fluorescence assay (Invitrogen).

Next Generation Sequencing

An amplicon library was generated from 10 ng of DNA from each sample using the Ion Ampliseq Cancer Panel (Life Technologies). Formalin-fixed, paraffin-embedded cell pellets of the H2122 cell line diluted in the HL60 cell line were used as control. The 46 genes in the panel for detection of targetable mutations included: AKT1, BRAF, FGFR1, GNAS, IDH1, FGFR2, KRAS, NRAS, PIK3CA, MET, RET, EGFR, JAK2, MPL, PDGFRA, PTEN, TP53, FGFR3, FLT3, KIT, ERBB2, ABL1, HNF1A, HRAS, ATM, RB1, CDH1, SMAD4, STK11, ALK, SRC, SMARCB1, VHL, MLH1, CTNNB1, KDR, FBXW7, APC, CSF1R, NPM1, SMO, ERBB4, CDKN2A, NOTCH1, JAK3, and PTPN11. Primers for PCR amplification included the 190-primer pair pool provided by the vendor with an additional primer pair that was custom added to cover the ‘hotspot’ location on codon 17 of AKT1. Following PCR amplification of target sequences, barcodes were ligated to the amplicons using the Ion Xpress Barcode Adaptors Kit (Life Technologies). Library quantification was then performed using the Bioanalyzer High Sensitivity DNA Chip (Agilent Technologies, Santa Clara, CA). The library was diluted in nuclease-free water to obtain a final concentration of 16 pM. Emulsion PCR was performed manually using the Ion Xpress Template Kit (Life Technologies) followed by manual breaking of the emulsion to isolate the ion spheres (ISPs). The quality of the DNA following PCR was measured using the Qubit IonSphere Quality control kit (Life Technologies). Selective ISPs with DNA were isolated and sequenced on a Ion 316 Chip (4 samples/chip) or a Ion 318 Chip (8 samples/chip) using the vendor-provided sequencing kit (Life Technologies). Successful sequencing of a sample required at least 300 000 reads with a quality score of AQ20 (1 misaligned base per 100 bases). For a wild-type call, a minimum coverage of 250× was required. As tumor specimens were admixed with normal tissue, a minimum coverage of 500× with at least 10% frequency was used as cutoff for a variant to be considered true. All variants detected by Ion PGM with at least 10% frequency were selected for confirmation by alternate platforms. Further details regarding the methodology and analysis have been described previously [12]. NGS performed by Foundation Medicine used the Illumina Hiseq 2000 Platform for 236 targetable GA's and the methods for the same have been described previously [13], [14].

Immunohistochemistry for HER2/Neu

A standard technique for fixing tissue in formalin and embedding it in paraffin was used. The 4-µm-thick histologic sections obtained from the TMAs were deparaffinized and hydrated in decreasing alcohol concentrations. Antigens were recovered by exposure to microwaves in citrate buffer (pH 6.0) and washed in PBS (pH 7.4). The monoclonal antibody anti-ErbB2 (NCL-CB11; Novocastra) was used at a dilution of 1∶40. The primary antibody was incubated at room temperature for 60 minutes and then incubated with the complex Super Picture Polymer Detection Kit (Zymed) in a Dako autostainer. Standard criteria for HER2/Neu scoring were utilized [15].

Statistical Analysis

We investigated potential associations between genetic mutations and survival in this cohort. All patients who had at least 3 months of follow up were included for the survival analysis. Overall survival (OS) was the chosen primary endpoint for this analysis. Progression-free survival (PFS) was a secondary endpoint as in prior clinical trials in CCA, PFS correlated accurately with overall survival (OS) [16]. Patient characteristics, age, gender, primary tumor site (intra- and extrahepatic), treatments administered, GA's observed, disease course, progression date and current status were recorded. The PFS and OS were estimated by using the Kaplan–Meier method. For survival analysis, the log rank test was used for discrete variables and the Cox proportional hazard model was used for continuous variables. All variables that were significant in univariate analysis and variables of our interest were entered into a multivariate model. The Cox proportional hazards model was used to calculate hazard ratio and 95% confidence intervals. A p-value of less than 0.05 was considered significant.

Ingenuity Pathway Analysis

The web-based pathways analysis tool IPA (Ingenuity Systems, www.ingenuity.com) was used to identify signaling pathways affected in CCA. This web-based entry tool allows for the mapping of gene expression data into relevant pathways based on their functional annotation and known molecular interactions. This knowledge coming from published, peer-reviewed scientific publications is stored in the Ingenuity Pathways Knowledge Base (IPKB), and is continuously updated. The mutated genes were uploaded into IPA. Canonical pathways analysis identified molecular pathways from the IPA library of canonical pathways (part of the IPKB) that were most significant to the data set. Genes from the data set that were associated with a canonical pathway in the IPKB were considered for the analysis. The significance of the association between the genes from the dataset and the canonical pathway (in the IPKB) was measured. Fisher's exact test was used to calculate a p-value determining the probability that there is an association between the genes in the dataset and the canonical pathway that cannot be explained by chance alone.

Intrahepatic CCA

Median age of the 55 patients with intrahepatic CCA was 60 years (range 24–84). Patient demographics are described in Table 2. The median PFS for these patients was 6.1 months and median OS was 10.9 months.

162 GAs were identified from 55 intrahepatic CCA patient samples with an average of 2.95 GAs/patient (range 0–7). GAs identified were mutations (66%), amplifications (20%), loss/deletions (7%) and others (7%). Only 4 (7%) tumors showed no detectable GAs using the NGS. Most frequent GAs were TP53 (35%), KRAS (24%), ARID1A (20%), IDH1 (18%), MCL1 (16%) and PBRM1 (11%). FGF pathway GAs occurred in 13% of cases, these included mutations, amplifications and FGFR fusion genes. The latter were noted in 3 cases: FGFR2-KIAA1598, FGFR2-NOL4 and FGFR2-PARK2 fusions. A schematic representation of the FGFR2-KIAA1598 fusion gene is depicted in Fig. 3. Signaling pathways associated with these mutations are depicted in Table 3. MAPK, chromatin modification and mTOR pathway aberrations were relatively common (35%, 33% and 25%, respectively) followed by DNA repair (16%) and FGF signaling (13%).

Potential associations between the mutations with PFS and overall survival OS were explored in this study. Multivariate regression analyses for PFS and OS are described in Table 4 and Table 5 respectively. In intrahepatic CCA, PFS was significantly associated with the clinical stage, KRAS mutations, and aberrations in the MAP/ERK pathway. The median PFS for patients with and without KRAS mutations was 3.4 and 10 months (p = 0.01). Median PFS for patients with and without aberrant MAP/ERK pathway genes was 3.9 and 10 months (p = 0.004). PFS was not significantly associated with age, gender, ethnicity, or presence of other mutations including IDH1.

The median OS for stage II, III and IV was 20.5, 15.6 and 6 months respectively (p = 0.017). OS was significantly associated with clinical stage, surgery, gender, TP53 mutations, KRAS mutations, MAP/ERK, mTOR and FGF pathway GAs (Table 6). Kaplan-Meier curves showing the relationship between OS and GA's in KRAS, TP53 and FGF genes are depicted in Figs. 4, 5, and 6, respectively. A multivariate model was set up to investigate the effect of the clinical variables and GAs on PFS and OS. No factor was found to be significantly associated with PFS on multivariate analysis. However, male gender (HR 3.25, p = 0.01) and TP53 mutations (HR 4.92, p = 0.0006) were found to be associated with reduced OS.

Extrahepatic CCA

Similar analyses were conducted for cases of extrahepatic CCAs. Median age of the 20 patients with extrahepatic CCA was 62.5 years (range 33–80). Patient demographics for this group are also described in Table 2. The median PFS for these patients was 7.7 months and median OS was 14 months.

A total of 78 GAs were identified from 20 extrahepatic CCA patient samples with an average of 3.9 GAs/patient (range 0–10). GAs identified were mutations (77%), amplifications (6%), loss/deletions (10%) and splices (6%). Only 1 tumor showed no detectable GA. Most frequent GAs were TP53 (45%), KRAS (40%), ERBB2 (25%), SMAD4 (25%), FBXW7 (15%), CDKN2A (15%) and CDKN2B (15%). ERBB2 GAs included 4 mutations and 1 case of amplification. One of the ERBB2 mutations was in the kinase domain (V777L) while 3 involved the extracellular domain (S310F). IHC for ERBB2 was negative in all cases with ERBB2 mutations. Signaling pathways associated with mutations in extrahepatic CCAs are depicted in Table 3. MAPK, mTOR and DNA repair pathway aberrations were relatively common (55%, 40% and 45%, respectively) followed by chromatin modification (15%) and FGF signaling (5%).

Survival analysis of extrahepatic CCA cases noted significant correlation of PFS with surgery, radiation, BAP1 mutation, and FGF pathway aberrations. OS was found to be associated with clinical stage, surgery, radiation therapy, and mutations in PBRM1 and BAP1 genes (Table 7). A Kaplan-Meier curve demonstrating the relationship between BAP1 mutations and overall survival are shown in Fig. 7. The presence of BAP1 mutations was significantly associated with reduced PFS (median: 3 vs. 8.8 months, p = 0.02) and reduced OS (median: 8.9 vs. 19.9 months, p = 0.007).

BAP1 Mutations

In our series, 6 cases had BAP1 mutations; 3 of these experienced early recurrence (less than six months) after potentially curative surgical resection. Bone and soft tissue metastases were common in this subpopulation. One such case is illustrated in Fig. 8; this patient experienced disease recurrence 7 weeks after surgical resection. In addition, all cases with BAP1 mutation were treated with systemic chemotherapy consisting of gemcitabine and cisplatin. No radiologic response was noted and progression after first line chemotherapy occurred in all cases within a short time frame (less than 4 months). Despite the limited number of cases, the aggressive clinical prognosis of this subgroup was very evident.

FGFR Pathway

Patients with FGFR pathway GAs appeared to have a relatively indolent prognosis in our series. FGFR Pathway GAs included mutations, fusion genes and ligand amplifications. Three patients in this subgroup had PFS>60 weeks with first line chemotherapy. These patients had FGFR2 mutation, FGFR2-PARK2 fusion and FGF19 amplification, respectively. Given the favorable prognosis of these cases, one patient with sustained response to systemic chemotherapy lasting for >2 years, received orthotopic liver transplantation with curative intent. In contrast, one patient with concomitant FGFR2-NOL4 fusion had a rapidly progressive course; this patient also had co-existing BAP1 mutation.

Response to Targeted Therapy

Based on their mutational profiles, patients were referred to appropriate clinical trials. Twelve patients were enrolled in phase I or II clinical trials. Clinical outcome from targeted therapy is illustrated below: Sustained stable disease was observed in 7 cases with KRAS wt on erlotinib. Two cases with sustained stable disease are illustrated in Fig. 9. One patient with KRAS mutation was enrolled on a clinical trial with pazopanib + trametinib and experienced stable disease after prior progression on first-line chemotherapy as illustrated in Fig. 10. Two patients with BRAF mutation were referred for a trial with BRAF inhibitor; both experienced partial responses. One of these cases with a metabolic response on FDG-PET Scan is illustrated in Fig. 11; this patient continues on therapy for 5+ months at this time. Another patient with c-met mutation was enrolled on an expansion cohort with a c-met inhibitor and experienced a metabolic response as illustrated in Fig. 12. In case of her2/neu mutations, no response was observed in two cases treated with trastuzumab and lapatinib, respectively.

These cases illustrate that the mutational profile not only provided prognostic information but also provided therapeutic options.