Noninvasive Molecular Diagnosis Using Circulating DNA Sequencing Improves Outcomes in Cholangiocarcinoma Management

TOP - November 2019, Vol 12, No 4 - Cholangiocarcinoma
Diagnosis of cholangiocarcinoma (CCA) is often identified at a late stage. Analyzing the tumor­specific mutation profile of a patient with CCA can improve the diagnosis and treatment for the individual patient. The molecular profile of CCA can be done through the use of circulating tumor (ct) DNA sequencing, which may help to target specific mutations and improve treatment selection for this rare type of cancer.

Based on the results of a recent study, molecular analysis of tumor tissue and blood samples before and during chemotherapy in patients with locally advanced or metastatic CCA reveals a high concordance rate between blood and tissue.1 This study was led by Thomas J. Ettrich, MD, Head, Outpatient Clinic and Clinical Trial Office in GI-Oncology, Ulm University Hospital, Department of Internal Medicine, Germany.

The mutation profile of the blood–tumor pairs was concordant for approximately 75% of the patients, the results showed. In the subgroup with intra­hepatic CCA (IHCC), the concordance rate exceeded 90%, highlighting the importance of molecular profiling in the management of patients with CCA and the potential role for targeted therapies.

Few patients with CCA—including IHCC and extrahepatic CCA (EHCC), as well as gallbladder cancer—are diagnosed at a resectable disease stage, and disease relapse rates after curative surgery are high.

Currently, cisplatin plus gemcita­bine combination chemotherapy is generally used as a first-line palliative treatment regimen. No treatment regimen for later lines of therapy has been established as the standard of care, and no targeted therapies are yet approved by the FDA for CCA, although several targeted therapies are currently being investigated.

Study Details

A total of 32 patients with histologically confirmed locally advanced or metastatic (stage III and IV) CCA were enrolled in the study. Patients with gallbladder cancer were excluded. Blood samples for ctDNA analysis were collected before treatment initiation (ie, baseline), at a median of 1.7 months after treatment initiation, and at disease progression.

In the original cohort of 24 patients, next-generation sequencing (NGS) was performed with a 15-gene panel consisting of the most frequent genetic mutations identified in CCA, including TP53, KRAS, ARID1A, BAP1, PBRM1, PIK3CA, SMAD4, FBXW7, IDH1, BCL2, BRAF, CDKN2A, ERBB2, IDH2, and NRAS.

To identify mutation signatures indicating disease progression, the researchers included ctDNA samples from baseline, during treatment, and at disease progression of an additional 8 patients as part of an expanded targeted panel resequencing comprising 710 cancer-­related genetic mutations.

In the original cohort of 24 patients, 13 (54%) tumors were categorized as IHCC and 11 (46%) as EHCC. Tumor biopsy material of 23 of the patients was analyzed by targeted enrichment and NGS of 15 genes; and in 9 of these genes, 22 mutations were found. Some 61% of patients had at least 1 mutation in tumor-driving genes.

Genetic Mutations in CCA

The overall concordance rate between ctDNA blood and tissue results was 74%. For patients with IHCC, the concordance rate was 92%, and for patients with EHCC, 55%.

“The lower concordance (55%) in EHCC may be due to specific features of the anatomic site, but also underlines that IHCC and EHCC may indeed be distinct diseases, with differences in genomic pattern and clinical behavior,”1 the authors noted.

The number of unique variants per gene was not significantly different between tumor tissue and ctDNA (P = .3125). The largest number of unique variants was detected in the TP53 gene (22% in tumor tissue, 30% by ctDNA), followed by KRAS (22% and 9%, respectively).

Overall, 67% of mutations were concordant between tumor tissue and ctDNA. Six (25%) tumor mutations could not be detected in ctDNA, whereas 2 (8%) mutations found in ctDNA were not detected in the paired tumor sample.

Prognosis

Some mutations detected in ctDNA have prognostic value, noted the authors. For example, mutations in BAP1, PBRM1, KRAS, and TP53 have been associated in previous studies with poor overall survival, even in the curative disease stage. Similarly, the current study demonstrated a trend toward shorter progression-free survival in patients with a mutation in 1 of these genes compared with patients without a mutation in these genes.

Changes in the molecular constitution of CCA were detected during chemotherapy. The investigators identified 149 genetic mutations in patients with CCA before and during chemotherapy, as well as a group of 76 genes with variants that emerged during chemotherapy. These genetic mutations likely drove the disease progression, “suggesting that tumor progress in CCA is a rather heterogenous process and large-scale panels are needed to monitor treatment-induced tumor evolution in this disease,”1 the researchers wrote.

The most frequent gene mutations found at disease progression were ERBB2, KMT2C, MUC1, ARID1A, CBLB, FOXE1, GATA6, and MAP3K4. Except for ARID1A, “all these genes do not belong to the assumed and published most frequently mutated genes in CCA,”1 Dr Ettrich and colleagues added.

Practical Implications

Mutations in TP53 and KRAS were significantly associated with worse survival and are more common in patients with EHCC, whereas mutations and fusions in FGFR1-3, IDH1/2, and ­ARID1A were more common in patients with IHCC.

“These specific differences in the mutational landscape open for separate therapeutic targeting for example with the dual BCR/ABL and Src family tyrosine kinase inhibitor dasatinib (NCT02428855) or AG120 and IDH1-inhibitors,”1 Dr Ettrich and colleagues noted.

“ctDNA sequencing harbors great potential to improve the clinical management of CCA patients. Mutations detected in ctDNA are representative for the respective tumor tissue (especially for IHCC), paving the way to a non-invasive molecular diagnosis and therapy stratification. Our data are encouraging for the estimation of the individual tumor load and the expected prognosis, which also might influence treatment decisions,”1 the researchers concluded.

Reference

  1. Ettrich TJ, Schwerdel D, Dolnik A, et al. Genotyping of circulating tumor DNA in cholangiocarcinoma reveals diagnostic and prognostic information. Sci Rep. 2019;9(1):13261. doi.org/10.1038/s41598-019-49860-0.
Last modified: November 8, 2019