Cholangiocarcinoma (CCA) is an aggressive cancer often diagnosed at an advanced stage and is associated with a poor prognosis. CCA has a median overall survival of approximately 1 year, even with the addition of durvalumab or pembrolizumab to gemcitabine and cisplatin as first-line therapy.1,2
Targetable genomic alterations in several genes have been identified, including FGFR2, BRAF, IDH1, RET, ERBB2, and NTRK, and have increased the availability of therapeutic options for patients with advanced CCA (aCCA). Due to aCCA tissue biopsies being associated with failure rates approaching 26.8%, liquid biopsy has become an important tool for identifying actionable molecular alterations.3
Amit Mahipal, MD, presented results for a real-world data (RWD) study to examine the rates of molecular alterations detected using circulating tumor DNA (ctDNA). Additionally, overall treatment patterns in aCCA were analyzed, and outcomes were assessed for patients receiving ivosidenib following ctDNA-detected IDH1 mutations.
RWD was obtained from the GuardantINFORM (Guardant Health) database, comprising aggregated commercial health claims and deidentified records from >350,000 patients with clinical ctDNA testing performed using Guardant 360 from 2014 to 2023.
Patients with aCCA were included in this analysis if they had ≥1 ctDNA alterations and >1 treatment claim. Patients were included in a subgroup analysis for intrahepatic CCA (iCCA) if they had ≥2 diagnosis codes for iCCA and <2 for extrahepatic CCA. Patient outcomes were assessed in months using real-world time to treatment discontinuation (rwTTD), real-world time to next treatment (rwTTNT), and real-world overall survival (rwOS).
There were 1495 patients with aCCA, and at least 1 genomic alteration was identified using the GuardantINFORM database. In total, 3.4% of patients had ctDNA testing before first-line therapy, 45% had ctDNA testing after first-line therapy, and 22% had ctDNA testing after second-line therapy.
In patients with aCCA, TP53 was the most frequently altered gene (56%), followed by KRAS (21%), ATM (16%), ARID1A (14%), FGFR2 (13%), EGFR (11%), PIK3CA (11%), BRCA2 (7%), and SMAD4 (7%).
In 403 patients with iCCA, 10.9% had an IDH1 mutation, 9.2% had an FGFR2 fusion, 1.2% had MSI-H detected, 1.2% had BRAFV600E mutation, <1% had an ERBB2 amplification, and <1% had a RET fusion. These rates are consistent with reported rates for tissue testing in the literature.4
The most common first-line therapy consisted of gemcitabine and cisplatin (49%). The most common second-line therapies consisted of a combination of leucovorin, fluorouracil, and oxaliplatin (21%), gemcitabine and cisplatin (15%), and capecitabine (11%). Among ivosidenib-naïve patients with an IDH1 mutation, 58 (20%) started therapy within 90 days of the ctDNA test, while 36% received ivosidenib, 57% received chemotherapy, and 7% received other therapy.
Although it did not reach statistical significance, patients with IDH1 mutations treated with ivosidenib had a trend toward improved rwTTD and rwTTNT.
In patients with IDH1 mutations treated with ivosidenib, 21 had numerically improved rwTTD and rwTTNT compared with 33 who received chemotherapy: rwTTD, 4.6 months; 95% confidence interval (CI), 2.6-8.6, versus rwTTD, 2.8 months; 95% CI, 2.0-6.3, P=.10; rwTTNT, 11.0 months; 95% CI, 5.7-not estimable (NE), versus rwTTNT, 5.2 months; 95% CI, 3.3-NE, P=.26. There was no difference in rwOS between patients who received it compared with those who did not; however, the sample size was small.
Using RWD, it was demonstrated that patients with aCCA often receive ctDNA testing before starting first- or second-line therapy.
Although it did not reach statistical significance, patients with IDH1 mutations treated with ivosidenib had a trend toward improved rwTTD and rwTTNT. The data discussed here support the clinical utility of liquid biopsy to identify aCCA patients who may benefit from targeted therapy.5
To sign up for our newsletter or print publications, please enter your contact information below.