Systemic therapy for advanced cholangiocarcinoma

INTRODUCTION — Cholangiocarcinomas are rare malignancies arising from the epithelial cells of the intrahepatic and extrahepatic bile ducts. Cholangiocarcinoma is an aggressive tumor that can metastasize beyond the bile ducts to other intrahepatic locations, the peritoneum, and distant extrahepatic organs. (See "Clinical manifestations and diagnosis of cholangiocarcinoma".)

The approach to systemic therapy for advanced and metastatic cholangiocarcinoma will be reviewed here. Adjuvant and neoadjuvant therapy for localized cholangiocarcinoma, treatment of locally advanced unresectable nonmetastatic cholangiocarcinoma, systemic therapy for ampullary cancer, and systemic therapy for gallbladder cancer are discussed separately.

●(See "Adjuvant and neoadjuvant therapy for localized cholangiocarcinoma".)

●(See "Treatment options for locally advanced, unresectable, but nonmetastatic cholangiocarcinoma".)

●(See "Ampullary carcinoma: Treatment and prognosis".)

●(See "Treatment of advanced, unresectable gallbladder cancer".)

INITIAL THERAPY — The approach to initial systemic therapy for advanced biliary tract cancer is evolving. Chemotherapy is an integral part of initial therapy [1], but no chemotherapy regimen alone consistently extends survival beyond 8 to 15 months. Therefore, we encourage enrollment in clinical trials, where available. Of note, many clinical trials often enroll patients with different histologies such as cholangiocarcinoma, gallbladder cancer, and ampullary cancer. Although they arise in similar locations, these cancers all have a unique natural history and response to therapy [2]. For patients who are ineligible for or decline clinical trials, our approach to initial therapy is based on patient performance status and bilirubin levels.

Good performance status and no hyperbilirubinemia — For patients with advanced or metastatic cholangiocarcinoma who have good ECOG performance status (less than 2) (table 1) and a no hyperbilirubinemia (total bilirubin less than 2.5 times the upper limit of normal [ULN]), we suggest the addition of immunotherapy (either durvalumab or pembrolizumab) to gemcitabine plus cisplatin as initial treatment, as these regimens improved overall survival (OS) and demonstrated durable responses in separate randomized trials [3,4]. Although either chemoimmunotherapy regimen is appropriate, gemcitabine plus cisplatin and durvalumab has the advantage of a simpler maintenance regimen with single-agent durvalumab every four weeks, which may be preferred by some patients and providers. (See 'Gemcitabine plus cisplatin and durvalumab' below and 'Gemcitabine plus cisplatin and pembrolizumab' below.)

For patients who decline chemoimmunotherapy or are anticipated to not tolerate its toxicities, gemcitabine plus cisplatin (table 2) is an appropriate alternative, given combination chemoimmunotherapy confers a modest OS benefit. (See 'Gemcitabine plus cisplatin' below.)

Other options for initial therapy include gemcitabine plus oxaliplatin, gemcitabine plus capecitabine, gemcitabine plus nabpaclitaxel, and gemcitabine plus S-1 (where available). (See 'Other regimens' below.)

Gemcitabine plus cisplatin and durvalumab — The addition of durvalumab to gemcitabine plus cisplatin improves OS and has durable responses in patients with treatment-naïve advanced or metastatic biliary tract tumors and no hyperbilirubinemia (ie, total bilirubin less than 2.5 times the ULN). After completion of induction chemoimmunotherapy, maintenance therapy consists of single-agent durvalumab every four weeks until disease progression or unacceptable toxicity.

Gemcitabine plus cisplatin and durvalumab was investigated in a double-blind, placebo-controlled phase III trial (TOPAZ-1) of 685 patients with previously untreated, unresectable locally advanced or metastatic biliary tract cancer. Patients had either intrahepatic cholangiocarcinoma (56 percent), extrahepatic cholangiocarcinoma (19 percent), or gallbladder cancer (25 percent) [3]. In this study, patients were randomly assigned to receive either durvalumab (1500 mg every three weeks) or placebo in combination with gemcitabine (1000 mg/m²) and cisplatin (25 mg/m²) on days 1 and 8 every three weeks for up to eight cycles. Patients subsequently received maintenance therapy with either durvalumab (1500 mg every four weeks) or placebo until disease progression or unacceptable toxicity.

At median follow-up of 17 months, compared with placebo plus chemotherapy, durvalumab plus chemotherapy improved OS (median 12.8 versus 11.5 months, HR 0.8, 95% CI 0.66-0.97), progression-free survival (median 7.2 versus 5.7 months, two-year OS 25 versus 10 percent, hazard ratio [HR] 0.75, 95% CI 0.63-0.89), and objective response rate (27 versus 19 percent). OS benefits were generally seen across clinically relevant subgroups including those with intrahepatic cholangiocarcinoma (HR 0.76, 95% CI 0.58-0.98), extrahepatic cholangiocarcinoma (HR 0.76), gallbladder cancer (HR 0.94), and PD-L1 expression: total area positivity ≥1 (HR 0.79, 95% CI 0.61-1.00). The proportion of ongoing responses lasting one year or longer for durvalumab versus placebo were 26 and 15 percent, respectively.

Combining durvalumab with chemotherapy also did not increase toxicity compared to placebo with chemotherapy (grade 3 to 4 toxicity rates of 76 versus 78 percent). (See "Toxicities associated with immune checkpoint inhibitors".)

Based on these data, the US Food and Drug Administration (FDA) has approved durvalumab, in combination with gemcitabine and cisplatin, for adult patients with locally advanced or metastatic biliary tract cancers [5].

Gemcitabine plus cisplatin and pembrolizumab — The addition of pembrolizumab to gemcitabine plus cisplatin is an option for initial therapy in patients with advanced or metastatic biliary tract tumors. In a phase III trial, this combination improved OS with durable responses and was well-tolerated [4]. After completion of induction chemoimmunotherapy, maintenance therapy for most patients consisted of both gemcitabine and pembrolizumab for up to 35 cycles (two years), including induction cycles, followed by gemcitabine monotherapy until disease progression or unacceptable toxicity.

Gemcitabine plus cisplatin and pembrolizumab was evaluated in a double-blind placebo-controlled phase III trial (KEYNOTE-966) of 1069 patients with previously untreated locally advanced or metastatic biliary tract cancer, including intrahepatic (59 percent) or extrahepatic cholangiocarcinoma (19 percent) and gallbladder cancer (22 percent) [4]. In this study, patients were randomly assigned to either pembrolizumab (200 mg every three weeks) or placebo, in combination with gemcitabine (1000 mg/m² on days 1 and 8 every three weeks) plus cisplatin (25 mg/m² on days 1 and 8 every three weeks) for up to eight cycles. For maintenance therapy, gemcitabine could be continued with no maximum duration and pembrolizumab or placebo was continued for up to 35 cumulative cycles (two years), or until disease progression or unacceptable toxicity. Patients who discontinued gemcitabine, cisplatin, or both due to toxicity could continue pembrolizumab or placebo, and vice-versa.

At median follow-up of 26 months, compared with placebo plus chemotherapy, pembrolizumab plus chemotherapy improved OS (median OS 12.7 versus 10.9 months, two-year OS 25 versus 18 percent, HR 0.83, 95% CI 0.72-0.95). Progression-free survival (median 6.5 versus 5.6 months, HR 0.86, 95% CI 0.75-1.00) and objective response rates (29 percent each) were similar between the treatment arms. OS benefits were generally seen across all clinically relevant subgroups, including those with intrahepatic cholangiocarcinoma (HR 0.76, 95% CI 0.64-0.91), extrahepatic cholangiocarcinoma (HR 0.99), and gallbladder cancer (HR 0.96), and PD-L1 combined positive score (CPS) ≥1 (HR 0.85, 95% CI 0.72-1.00). The proportion of ongoing responses at two years or longer for pembrolizumab versus placebo were 18 and 6 percent, respectively.

Combining pembrolizumab with chemotherapy also did not significantly increase toxicity compared with placebo plus chemotherapy (grade 3 to 4 toxicity of 79 versus 75 percent). (See "Toxicities associated with immune checkpoint inhibitors".)

Gemcitabine plus cisplatin and pembrolizumab is approved by the US FDA for the treatment of patients with locally advanced unresectable or metastatic biliary tract cancer (BTC) [6].

Gemcitabine plus cisplatin — Gemcitabine plus cisplatin (table 2) is an appropriate alternative for patients with treatment-naïve advanced cholangiocarcinoma, good ECOG performance status, and no hyperbilirubinemia (ie, total bilirubin less than 2.5 times the ULN) who decline chemoimmunotherapy or are anticipated to not tolerate its toxicities. Gemcitabine plus cisplatin improved OS over single-agent gemcitabine in a randomized trial [7], but it has not been directly compared with other gemcitabine-based combinations, except for gemcitabine plus S-1. (See 'Gemcitabine plus S-1' below.) Gemcitabine plus cisplatin is also well tolerated in most studies [8-13].

In initial clinical trials of advanced biliary tract cancers, gemcitabine demonstrated higher objective response rates and tumor control compared with regimens that did not include gemcitabine or cisplatin [13]. These data led to the conduct of an open-label phase III trial (ABC-02) of 410 patients with locally advanced or metastatic bile duct (59 percent), gallbladder (36 percent), or ampullary (5 percent) cancer [7]. In this study, patients were randomly assigned to 24 weeks of either gemcitabine (1000 mg/m²) plus cisplatin (25 mg/m²) on days 1 and 8 every 21 days, or gemcitabine alone (1000 mg/m² on days 1, 8, and 15 every 28 days).

At a median follow-up of approximately eight months, the combination improved OS (median 11.7 versus 8.1 months, HR 0.64, 95% CI 0.52-0.80) and PFS (median 8 versus 5 months, HR 0.63, 95% CI 0.51-0.77) compared with gemcitabine alone. Toxicity was similar in both treatment arms, except for higher rates of grade 3 or 4 neutropenia with gemcitabine plus cisplatin (25 versus 17 percent) and grade 3 or 4 abnormal liver function with gemcitabine alone (27 versus 17 percent). Most quality of life scales showed a trend favoring combined therapy, although the differences were not statistically significant [14]. Of the 21 long-term (>3 years) survivors, two-thirds (14 survivors) had received gemcitabine plus cisplatin.

Similar results were also seen in a separate randomized trial conducted in Japan, which also demonstrated an overall survival benefit for the combination over single-agent gemcitabine (11.2 versus 7.7 months) [11].

Other regimens — Other regimens are also available for initial treatment of advanced cholangiocarcinoma.

Gemcitabine plus oxaliplatin — Gemcitabine plus oxaliplatin (GEMOX) is an alternative option for initial therapy in advanced cholangiocarcinoma. GEMOX is used by some institutions since it is an active, well-tolerated regimen that was non-inferior to CAPOX in a phase III trial [15]. However, GEMOX has not been directly compared with gemcitabine plus cisplatin in randomized trials. In addition, initial therapy with GEMOX precludes the option of second-line therapy with fluorouracil plus leucovorin and oxaliplatin (FOLFOX) due to prior oxaliplatin exposure. (See 'Patients initially treated with gemcitabine plus oxaliplatin' below.)

Based on initial data from observational studies and early phase clinical trials [16-20], a randomized open-label phase III, noninferiority trial was conducted comparing GEMOX to capecitabine plus oxaliplatin (CAPOX) in 222 patients with treatment-naive advanced biliary tract cancer. In this study, GEMOX was non-inferior to CAPOX for OS (median 10.4 versus 10.6 months) and PFS (six-month PFS 45 versus 47 percent) [15]. Objective response rates were higher for GEMOX compared with CAPOX, but the difference was not statistically significant (25 versus 16 percent). Grade ≥3 toxicity rates were similar between the two treatment arms. Common grade 3 to 4 toxicities for GEMOX included neutropenia (14 percent) and thrombocytopenia (11 percent), and there were two cases of febrile neutropenia (1 percent).

The addition of bevacizumab to GEMOX has been investigated in clinical trials [21], but this approach remains investigational.

Gemcitabine plus capecitabine — Gemcitabine plus capecitabine is an active, well-tolerated regimen in advanced biliary tumors, based on data from phase II trials [22-27]. Although gemcitabine plus capecitabine is a reasonable alternative for initial therapy, it has not been directly compared with gemcitabine plus cisplatin in randomized trials.

●In a single-arm phase II trial, 45 patients with advanced cholangiocarcinoma or gallbladder cancer were treated with gemcitabine (1000 mg/m² on days 1 and 8) plus capecitabine (650 mg/m² twice daily for 14 days of every 21-day cycle) [22]. The objective response rate was 31 percent, including two complete responses (one for each tumor type). Median PFS and OS were 7 and 14 months, respectively.

●In a separate phase II trial conducted by the Southwest Oncology Group (SWOG) of 57 patients with advanced biliary tract cancer, gemcitabine plus capecitabine demonstrated an objective response rate of 25 percent, but median OS was 7 months [27].

Gemcitabine plus nabpaclitaxel — Gemcitabine plus nanoparticle albumin-bound paclitaxel (nabpaclitaxel (table 3)) is also an active initial regimen in advanced biliary tumors. This regimen is an appropriate alternative for those with a contraindication to, or without access to (table 4), cisplatin. However, this combination has not been directly compared with gemcitabine plus cisplatin in randomized trials. In a single-arm phase II trial of 74 patients with advanced or metastatic cholangiocarcinoma, gemcitabine plus nabpaclitaxel had an objective response rate was 30 percent [28]. Median PFS and OS were 8 and 12 months, respectively. Common grade ≥3 toxicities included neutropenia (43 percent) and fatigue (14 percent).

Although the triplet combination of gemcitabine plus cisplatin and nabpaclitaxel is also active in advanced cholangiocarcinoma, we do not use this regimen due to risk of increased toxicity and availability of other better-tolerated triplet therapies (ie, chemoimmunotherapy). Gemcitabine plus cisplatin and nabpaclitaxel was investigated in a phase II study of 60 patients with locally advanced unresectable or metastatic biliary tract cancer, including gallbladder and cholangiocarcinoma [29]. At median follow-up of 12 months, the objective response rate was 45 percent. Median PFS and OS were 12 and 19 months, respectively. The grade ≥3 toxicity rate was 58 percent, and the most frequent was neutropenia (33 percent). In addition, nine patients (16 percent) withdrew from the trial because of adverse events.

Gemcitabine plus S-1 — Gemcitabine plus S-1 is an option for initial treatment of advanced cholangiocarcinoma that was non-inferior to gemcitabine plus cisplatin in a randomized phase III trial [30]. S-1, which is available in some countries outside of the United States, is an oral fluoropyrimidine that includes three different agents: ftorafur (tegafur), gimeracil (5-chloro-2,4-dihydropyridine, a potent inhibitor of the FU-metabolizing enzyme dihydropyridine dehydrogenase [DPD]), and oteracil (potassium oxonate, which inhibits phosphorylation of intestinal FU, thought responsible for treatment-related diarrhea).

In a phase III non-inferiority trial (FUGA-BT) conducted in Japan, 354 patients with chemotherapy-naïve recurrent or unresectable adenocarcinoma of the gallbladder, biliary tract, or ampulla of Vater were randomly assigned to either gemcitabine (1000 mg/m² on days 1 and 8) plus S-1 (60, 80, or 100 mg daily based on body surface area and administered on days 1 to 14 of a 21-day cycle) or gemcitabine plus cisplatin (table 2) [30]. Relative to gemcitabine plus cisplatin, gemcitabine plus S-1 demonstrated noninferior OS (median 15.1 versus 13.4 months, HR 0.95, 95% CI 0.78-1.15), PFS (median 6.8 versus 5.8 months, HR 0.86, 95% CI 0.70-1.07), and objective response rate (30 versus 32 percent). Both treatments were generally well tolerated, although clinically relevant adverse effects (grade 2 or worse fatigue, anorexia, nausea, vomiting, mucositis, and diarrhea) were slightly more frequent with gemcitabine plus cisplatin (35 versus 31 percent).

Non-gemcitabine-based regimens — We prefer a gemcitabine-containing regimen as initial therapy for patients with advanced cholangiocarcinoma and no hyperbilirubinemia (ie, total bilirubin less than 2.5 times the ULN). However, data from randomized trials are insufficient to conclude that gemcitabine-based combinations are superior to non-gemcitabine-based regimens.

●Various randomized trials have directly compared gemcitabine-containing with non-gemcitabine-containing chemotherapy or chemoradiotherapy for initial therapy of advanced biliary tract cancer [31-37]. A Cochrane analysis concluded that the evidence obtained from these low-quality trials was insufficient to prove whether or not gemcitabine-containing combinations are better than non-gemcitabine-containing combinations, and that additional studies were needed [38].

●In a subsequent randomized phase II trial (PRODIGE 38 AMEBICA) of 191 patients with advanced biliary tract cancer (mostly cholangiocarcinoma) modified FOLFIRINOX (table 5) failed to improve OS or PFS relative to gemcitabine plus cisplatin.

Good performance status and persistent biliary obstruction

Fluorouracil-based regimens — For patients with a good performance status who have hyperbilirubinemia despite stenting, we prefer a non-gemcitabine-based regimen, such as LV-modulated FU (table 6) or a fluoropyrimidine plus oxaliplatin such as FOLFOX or CAPOX.

In past studies, objective response rates for FU alone or FU-based combination therapies ranged from 0 to 34 percent, and median survival was typically less than six months. Many, but not all, more recent series using either infusional FU in combination regimens or LV-modulated FU report higher response rates and marginally longer survival (but still less than one year) [39-46].

Borderline performance status — For patients with a borderline performance status or extensive comorbidity, options for initial therapy include LV-modulated FU (table 6), single-agent capecitabine, and single-agent gemcitabine (table 7).

Leucovorin-modulated fluorouracil — Objective response rates with FU alone are low, and median survival is typically short (usually less than six months) [41]. Higher response rates are reported in many series using either infusional FU or LV-modulated FU, although whether this translates into better survival is unclear [42-49]. In one report of 28 patients with advanced biliary tract cancer, FU (375 mg/m² per day by bolus) plus LV (25 mg/m² per day) was given on days 1 through 5 every three to four weeks [44]. There were nine objective responses (32 percent), two of which were complete and lasted for 14 and 16 months, respectively. However, median survival for the entire group was only six months.

Regimens that combine short-term infusional FU with LV (eg, the de Gramont schedule (table 6)) are better tolerated than bolus regimens, and these are generally preferred, although they require central venous access.

Capecitabine — Capecitabine monotherapy is a reasonable alternative to LV-modulated FU for initial therapy in patients with a borderline performance status, although for unclear reasons, capecitabine as a single agent appears to be relatively less active for cholangiocarcinoma than for gallbladder cancer [50,51]. (See "Treatment of advanced, unresectable gallbladder cancer", section on 'Borderline performance status'.)

Gemcitabine alone — Although it is better tolerated than combination therapy, gemcitabine alone has been associated with inferior outcomes when compared with gemcitabine-based doublets [7,11,52,53]. Overall, objective response rates with gemcitabine alone range from 7 to 27 percent, but median survival is rarely longer than eight months [8,54,55].

Although randomized trials have not been carried out, the combination of FU/LV and gemcitabine does not appear to be substantially more active than gemcitabine alone [56-58].

Biweekly gemcitabine plus cisplatin — Another option is biweekly cisplatin plus gemcitabine, which is associated with a more favorable toxicity profile than standard cisplatin plus gemcitabine [59].

SECOND-LINE THERAPY AND BEYOND — There are few prospective trials comparing specific chemotherapy regimens in the second-line setting for advanced cholangiocarcinoma, and the selection of candidates for second-line therapy as well as the optimal regimen are not established. Targeted testing of advanced cholangiocarcinomas for mismatch repair deficiency (dMMR)/microsatellite instability (MSI) and for specific molecular alterations for which a targeted treatment might be available is indicated for those who might be eligible for molecularly targeted therapy or immunotherapy, preferably within the context of a clinical trial. (See 'Molecularly targeted therapy' below.)

Prognostic stratification — The optimal selection of candidates for second-line chemotherapy is not established. Three independent studies suggest that patients who have a good performance status (0 or 1 (table 1)), disease control with initial chemotherapy, a relatively low carbohydrate antigen 19-9 (CA 19-9) level, an absence of peritoneal carcinomatosis, and possibly, previous surgery on their primary tumor have the longest survival with second-line chemotherapy [60-62], but whether these characteristics predict chemotherapy responsiveness or more favorable biologic behavior is not clear. No particular regimen seems superior to any other, and the choice of second-line regimen is empiric [61].

Cytotoxic chemotherapy — Largely based on the ABC-06 trial, for most patients who have disease progression while receiving gemcitabine plus cisplatin and who retain an adequate performance status, in the absence of potentially actionable molecular targets, we suggest treatment with short-term infusional fluorouracil (FU) plus leucovorin (LV) and oxaliplatin (FOLFOX). Other conventional chemotherapy regimens that could be considered in this setting include gemcitabine plus oxaliplatin (GEMOX) with or without bevacizumab, capecitabine plus oxaliplatin (CAPOX), liposomal irinotecan plus leucovorin-modulated FU, or a fluoropyrimidine alone. (See 'Gemcitabine plus oxaliplatin' above.)

After failure of GEMOX, gemcitabine plus capecitabine, capecitabine plus cisplatin, or short-term infusional FU plus LV and irinotecan (FOLFIRI) with or without bevacizumab are appropriate options. For selected patients, second-line molecularly targeted therapy using erlotinib plus bevacizumab may be considered; however, this regimen is highly costly.

Patients initially treated with gemcitabine plus cisplatin

FOLFOX — Short-term infusional FU plus LV and oxaliplatin (FOLFOX, (table 8)) is an active regimen for second-line therapy [53,63,64]. FOLFOX was directly compared with active symptom control alone in the randomized ABC-06 trial, which enrolled 162 patients with disease progression after prior gemcitabine plus cisplatin; there were 117 patients with cholangiocarcinoma, 34 with gallbladder cancer, and 11 with ampullary cancer [65]. FOLFOX was associated with significantly better rates of overall survival at 6 (51 versus 36 percent) and 12 months (26 versus 11 percent), and significantly, albeit modestly, improved median overall survival (6.2 versus 5.3 months, hazard ratio [HR] 0.69, 95% CI 0.50-0.97). Grade 3 or 4 toxic events were reported in 59 percent of the group receiving FOLFOX (versus 37 percent of those undergoing symptom control alone), with fatigue and neutropenia being more frequent in the FOLFOX arm. Notably, the protocol allowed for an initial 20 percent reduction in fluorouracil dose for individuals older than 70, and a reduced initial oxaliplatin dose (65 rather than 85 mg/m²) if the creatinine clearance was 30 to 60 mL/min.

Liposomal irinotecan — A benefit for second-line liposomal irinotecan in combination with leucovorin [LV]-modulated short-term infusional FU (table 9) compared with FU/LV alone after progression on gemcitabine plus cisplatin was addressed in the phase II NIFTY trial [66]. Of the 174 patients who were assessable for response, median progression-free survival (the primary endpoint, as assessed by blinded independent central review) was significantly better in the irinotecan group (7.1 versus 1.4 months) as was median overall survival (8.6 versus 5.5 months) and objective response rate (15 versus 6 percent). The most common grade 3 or worse adverse events in the irinotecan group were neutropenia, fatigue, and nausea.

Capecitabine plus oxaliplatin — All of the data on CAPOX are in the initial setting:

●Benefit for CAPOX for initial therapy was also shown in a phase II trial of 65 patients with advanced biliary cancer (38 with intrahepatic or extrahepatic cholangiocarcinoma) [26]. Among the patients with extrahepatic bile duct cholangiocarcinoma, there were two complete and eight partial responses. There were no partial or complete responses among patients with intrahepatic mass-forming cholangiocarcinoma. Treatment was well tolerated, with only mild hematologic toxicity, grade 3 or 4 peripheral neuropathy in 11 patients of the entire cohort, and two hypersensitivity reactions to oxaliplatin.

●Others have demonstrated noninferiority for CAPOX compared with GEMOX as initial therapy for advanced biliary tract cancer [15].

We would not pursue CAPOX for a patient initially treated with GEMOX. However, second-line therapy with this combination may be an appropriate alternative to FOLFOX after failure of gemcitabine plus cisplatin.

Capecitabine plus irinotecan — Another second-line alternative after failing initial gemcitabine plus cisplatin is XELIRI. However, in our view, given the potential for treatment-related toxicity, other regimens such as liposomal irinotecan plus leucovorin-modulated FU are preferred.

XELIRI (irinotecan 180 mg/m² on day 1 plus capecitabine 1000 mg/m² twice daily on days 1 to 10 of every 14-day cycle) was directly compared with irinotecan alone (180 mg/m² on day 1 of each 14-day cycle) in a randomized phase II trial of 64 patients with advanced biliary tract cancer (12 with gallbladder cancer, the remainder with cholangiocarcinoma) previously treated with gemcitabine plus cisplatin [67]. Median progression-free survival (3.7 versus 2.4 months) and nine-month survival (61 versus 32 percent) were significantly better with combined therapy, although the difference in median overall survival was not significant (10.1 versus 7.3 months, p = 0.107). Rates of severe (grade 3 or 4) toxicity were not much worse with combined therapy, with the exception of palmar-plantar erythrodysesthesia (6.7 versus 0 percent).

Fluoropyrimidine alone — Whether results are better with FOLFOX or CAPOX than a fluoropyrimidine alone is unclear; there are no randomized trials. One retrospective analysis of 321 patients who received a fluoropyrimidine-based second-line regimen after failing gemcitabine plus cisplatin concluded that the objective response rate was higher with a fluoropyrimidine-platinum combination (8 versus 1 percent) but that progression-free and overall survival were not significantly better [68].

Antiangiogenic therapy — Angiogenesis is an important target in biliary tract cancers, as vascular endothelial growth factor (VEGF) and VEGF receptor (VEGFR) overexpression are reported in 30 to 50 percent of tumors and associated with a poor prognosis [69]. Unfortunately, the benefits in clinical trials have been modest, at best.

●Bevacizumab – The efficacy of bevacizumab, a monoclonal antibody targeting VEGF, in combination with a variety of other drugs has been promising in uncontrolled studies in metastatic biliary tract cancer [21,70-73], but the contribution of this agent remains uncertain due to the lack of controlled trials. Bevacizumab plus erlotinib is discussed below. (See 'What is the role of EGFR inhibitors?' below.)

●Regorafenib – Regorafenib is an orally active inhibitor of angiogenic (including the VEGFRs 1 to 3), stromal, and oncogenic receptor tyrosine kinases that targets a variety of kinases implicated in angiogenic and tumor growth-promoting pathways.

A benefit for regorafenib compared with best supportive care after failure of gemcitabine and platinum-based chemotherapy was noted in REACHIN, a randomized double-blind phase II trial [74]. Overall, 66 patients were randomly assigned to regorafenib (160 mg daily) or an identically matched placebo. Median progression-free survival, the primary endpoint, was modestly but significantly higher with regorafenib (3.0 versus 1.5 months), and the disease control rate was also higher (70 versus 33 percent); however, median overall survival was similar (5.3 versus 5.1 months). The most common treatment-related grade 3 or 4 toxicities with regorafenib were fatigue (18 percent), nausea and vomiting (9 percent), and skin toxicity (9 percent).

A similar modest degree of benefit was reported in another open label multi-institutional trial conducted in patients with heavily pretreated advanced biliary tract cancer (median progression-free survival 3.7 months, disease control rate 64 percent [75]).

Given these modest benefits, we would not pursue regorafenib as a second-line strategy but consider it an option for third-line therapy.

●Ramucirumab – Ramucirumab is a fully human IGG1 monoclonal antibody that targets VEGFR2, inhibiting the biding of VEGF. Activity in advanced pretreated biliary tract cancer was shown in a phase II study of 61 patients (62 percent intrahepatic cholangiocarcinoma, 17 percent extrahepatic, and the remainder gallbladder cancer); all had received gemcitabine plus cisplatin frontline, and eight had received four or more prior therapies [76]. One patient achieved a partial response, and the disease control rate overall was 45 percent. Median progression-free survival was 3.2 months, and median overall survival 9.5 months. Ramucirumab was well tolerated, and the majority of toxicities were grade 1 or 2.

Whether these results are better than can be achieved with any other salvage regimen in this setting is not clear; there are no comparator trials.

Patients initially treated with gemcitabine plus oxaliplatin

Gemcitabine plus capecitabine — The combination of gemcitabine plus capecitabine is active for advanced biliary tumors.

Capecitabine plus cisplatin — Capecitabine, as an alternative to infusional FU, has been investigated in combination with cisplatin. A relatively large (although retrospective) study of 176 patients reported a 17 percent response rate and median survival of 7.4 months [77]. Another smaller but prospective study achieved a similar 21 percent response rate and median survival of 9.1 months [78].

FOLFIRI with or without bevacizumab — The efficacy of second-line short-term infusional FU plus LV and irinotecan (FOLFIRI) plus bevacizumab was addressed in a small retrospective analysis of 13 patients who were refractory to initial therapy with GEMOX [70]. There were five objective responders (one complete), and median overall survival was 20 months.

Molecularly targeted therapy — Targeted testing of advanced cholangiocarcinomas for dMMR/MSI and for specific molecular alterations for which a targeted treatment might be available is indicated for those who might be eligible for molecularly targeted therapy or immunotherapy, preferably within the context of a clinical trial [79].

Next-generation sequencing to identify actionable molecular abnormalities — Several ongoing trials (eg, the National Cancer Institute [NCI] Molecular Analysis for Therapy Choice [MATCH] and the ASCO Targeted Agent and Profiling Utilization Registry [TAPUR] trails) are using next-generation sequencing (NGS) of multiple genes (gene panel tests) to identify molecular abnormalities in the tumors of patients with refractory cancers that may potentially match molecularly targeted therapies that are either in clinical trials or approved for treatment of other cancer types. Two such gene panel tests (MSK-IMPACT [Memorial Sloan Kettering Cancer Center Integrated Mutation Profiling of Actionable Cancer Targets] and F1CDx [FoundationOne CDx]) are US Food and Drug Administration (FDA) approved in the United States. These tests can be used on formalin-fixed, paraffin-embedded (FFPE) tissue regardless of the primary organ from which the tumor arose. (See "Next-generation DNA sequencing (NGS): Principles and clinical applications", section on 'Cancer screening and management'.)

Biliary tract carcinomas (particularly intrahepatic cholangiocarcinomas) have multiple molecular alterations, many of which are potential targets for available specific inhibitors [80-87]. For patients without a molecular alteration, or those with molecular alterations for which a targeted treatment is not available, clinical trial enrollment is encouraged, where available.

The potential impact of this approach in patients with advanced biliary tract cancer can be illustrated by the following reports:

●In the Molecular Screening for Cancer Treatment Optimization (MOSCATO)-01 trial, 43 of 1035 adults with advanced cancer had a biliary tract malignancy, 34 of whom successfully underwent high-throughput molecular screening on fresh frozen tissue [86]. Potentially actionable molecular aberrations were identified in 23 patients (68 percent), 18 of whom received targeted treatment. Median progression-free survival was 5.2 months, and there were six objective responses (33 percent, one complete).

Interestingly, all of the patients harboring a fusion transcript involving the fibroblast growth factor receptor 2 (FGFR2) and the single patient with an FGFR2 mutation had dramatic responses with therapeutic FGFR inhibition. Others have reported favorable results from the targeting of FGFR mutations/fusions in advanced cholangiocarcinoma [88-90]. These data are discussed below. (See 'FGFR inhibitors for FGFR fusion-positive tumors' below.)

The two best responders in this series had been treated with human epidermal growth factor 2 (HER2) inhibitors, one of whom had amplification of ERBB2 and the other had a mutation in ERBB3.

Besides these, other promising targets for biliary tract cancers include mutations in the isocitrate dehydrogenase (IDH) genes IDH1 and IDH2 [80,91], for which specific inhibitors are available (eg, ivosidenib, enasidenib), and the BRAF V600E mutation. (See 'Ivosidenib for IDH-mutated cholangiocarcinoma' below.)

●In a database series of 454 biliary tract cancers, potentially actionable molecular alterations were identified in 30.5 percent, including 39 percent of intrahepatic cholangiocarcinomas, 30 percent of extrahepatic cholangiocarcinomas, and 15 percent of gallbladder cancers [87]. This series used a narrower definition of potentially actionable alterations, and only included those findings for which a tumor-agnostic drug approval existed in the United States in 2022.

Immunotherapy — Immunotherapy with an immune checkpoint inhibitor may be particularly beneficial for patients whose tumors have DNA dMMR/high levels of MSI (MSI-H), programmed cell death 1 ligand 1 (PD-L1) overexpression, or high levels of tumor mutational burden (TMB-high); however, emerging data suggest that durable responses are also possible in patients with intrahepatic cholangiocarcinomas unselected for biomarker status. We would not typically offer second-line immunotherapy to individuals who received durvalumab or pembrolizumab as part of initial therapy.

Biomarker-selected patients

●dMMR/MSI-H – It has been hypothesized that tumors that lack the mismatch repair mechanism (ie, mismatch repair deficiency [dMMR]) harbor many more mutations (ie, they are hypermutated) than do tumors of the same type without such mismatch repair defects, and that the neoantigens generated from mutations such as these are more immunogenic than those generated by other mutations. The biologic footprint of dMMR tumors is high levels of MSI (MSI-H).

Several steps are required for the immune system to effectively attack tumor cells. Several immune checkpoints exist to dampen the immune response in order to protect against detrimental inflammation and autoimmunity. In the setting of malignancy, such immune checkpoints can result in immune tolerance of the tumor and subsequent progression of the malignancy. Inhibition of these checkpoints might be expected to halt/reverse disease progression. One well-characterized checkpoint being targeted in several tumor types is programmed cell death 1 (PD-1). PD-1 is upregulated on activated T cells, and upon recognition of tumor via the T cell receptor, PD-1 engagement by PD-L1 expressed by tumor or other immune cells infiltrating the tumor tissue can lead to T cell inactivation and a "brake" on immune-mediated tumor eradication. (See "Principles of cancer immunotherapy".)

Proof of principle that cancers with dMMR might be particularly susceptible to inhibition of the PD-L1/PD-1 interaction was initially provided by a study of pembrolizumab in dMMR colorectal cancer. Subsequently, it was established that immune checkpoint inhibitor immunotherapy appears to benefit a subset of patients with dMMR tumors regardless of anatomic site of origin or tissue histology. As a result, in May 2017, the FDA approved pembrolizumab for treatment of a variety of advanced solid tumors, including cholangiocarcinomas, that had MSI-H or dMMR, that had progressed following prior treatment, and for which there were no satisfactory alternative treatment options, the first such approval of a tissue-agnostic anticancer treatment. (See "Tissue-agnostic cancer therapy: DNA mismatch repair deficiency, tumor mutational burden, and response to immune checkpoint blockade in solid tumors", section on 'Other tumors with MSI-H/dMMR'.)

Efficacy in dMMR advanced cholangiocarcinoma was shown in the phase II KEYNOTE-158 study, which enrolled 22 patients with cholangiocarcinoma [92]. There were nine objective responses (41 percent), two of which were complete, and the median duration of response ranged from 4.1 to 24.9+ months.

The fraction of cholangiocarcinomas that have dMMR has been addressed in the following reports:

•In one study, approximately 3 percent of cholangiocarcinomas had dMMR or MSI-H [93].

•The frequency of dMMR or MSI-H in other studies has been approximately 5 percent each for gallbladder cancer and extrahepatic cholangiocarcinoma and 10 percent each for intrahepatic cholangiocarcinoma and ampullary carcinoma [94].

•Lower levels of dMMR (1 to 1.3 percent) have been reported for biliary tract cancers when the analysis was limited to individuals with advanced/metastatic disease [93,95].

The approach to testing for dMMR is addressed in detail elsewhere. (See "Tissue-agnostic cancer therapy: DNA mismatch repair deficiency, tumor mutational burden, and response to immune checkpoint blockade in solid tumors", section on 'Assessing mismatch repair' and "Tissue-agnostic cancer therapy: DNA mismatch repair deficiency, tumor mutational burden, and response to immune checkpoint blockade in solid tumors", section on 'Approach to testing dMMR as a predictive marker'.)

●PD-L1 overexpression – Other biomarkers such as overexpression of PD-L1 may also predict response to immune checkpoint inhibitor therapy. As an example, in one phase II trial of nivolumab in 46 patients with advanced refractory biliary tract, 18 of 42 tested tumors (43 percent) had PD-L1 overexpression. Overall, 10 patients achieved an investigator-assessed immune partial response (table 10), all of whom had tumors with proficient mismatch repair (pMMR) [96]; 9 of the 10 overexpressed PD-L1. Four responders achieved durable objective response lasting at least one year.

On the other hand, a combined analysis of data from the KEYNOTE-028 and 158 studies suggest that pembrolizumab monotherapy has limited activity in advanced biliary tract cancer, irrespective of PD-L1 overexpression [97].

Taken together, these limited data suggest that responses appear to be more frequent in patients with PD-L1 positive tumors, although they have also been documented with PD-L1 negative tumors.

Combined immunotherapy approaches may be more beneficial, at least for patients with intrahepatic cholangiocarcinoma, although the predictive value of PD-L1 overexpression remains uncertain. (See 'Biomarker-unselected patients' below.)

●Tumor mutational burden – Biliary tract cancers are generally characterized by low TMB, with only 2.9 to 4.0 percent of cases having high levels of TMB (which has been defined variably) [98-101]. Higher levels of TMB appear to correlate with benefit from immune checkpoint inhibitor immunotherapy in a variety of tumor types [101,102]. (See "Tissue-agnostic cancer therapy: DNA mismatch repair deficiency, tumor mutational burden, and response to immune checkpoint blockade in solid tumors", section on 'Tumors with high mutational burden'.)

There is scant information on responsiveness of TMB-high cholangiocarcinomas to immune checkpoint inhibitor therapy:

•A single published case report documents durable disease control with treatments targeting the PD-1 pathway in a single patient with TMB-high advanced biliary tract cancer [103].

•The KEYNOTE-158 trial, which assessed the benefit of pembrolizumab monotherapy in a variety of malignancies, including cholangiocarcinoma, reported a correlation between high TMB and objective response to pembrolizumab monotherapy, and responses were seen in anal, cervical, endometrial, salivary, thyroid, or vulvar carcinoma; mesothelioma; a neuroendocrine tumor; and small cell lung cancer [104]. However, none of the cholangiocarcinomas enrolled in the study had TMB-high disease; 2 of 63 patients without high TMB had an objective response.

However, largely based on this study, pembrolizumab is now approved for patients with any solid tumor, including cholangiocarcinoma, that has a TMB ≥10 mut/Mb (as defined by the FDA-approved companion diagnostic test), after progression on standard regimens. (See "Tissue-agnostic cancer therapy: DNA mismatch repair deficiency, tumor mutational burden, and response to immune checkpoint blockade in solid tumors", section on 'Tumors with high mutational burden'.)

Although it remains uncertain whether this is the appropriate threshold to define high TMB in biliary tract cancer, as validation studies were conducted mainly in lung and urothelial cancers and thresholds for TMB are likely to vary across tumor types, we suggest using the NGS platform FoundationOne CDx assay to assess TMB, as this was the platform used in the KEYNOTE-158 trial, and the use of a threshold TMB of ≥10 mut/Mb to select patients for immune checkpoint inhibitor immunotherapy. Cutoffs for TMB using other platforms have not been identified or linked to the FoundationOne data. (See "Tissue-agnostic cancer therapy: DNA mismatch repair deficiency, tumor mutational burden, and response to immune checkpoint blockade in solid tumors", section on 'Approach to testing for high levels of TMB'.)

Biomarker-unselected patients

●Combined immunotherapy for patients with intrahepatic cholangiocarcinoma – Until further information is available regarding the relevance of additional biomarkers on response rates to combined immunotherapy, a trial of nivolumab plus ipilimumab is reasonable in the setting of second-line therapy for patients with intrahepatic cholangiocarcinoma who do not have dMMR, high levels of TMB, or PD-L1 overexpression, and who did not receive frontline durvalumab.

Emerging data suggest that durable responses to immune checkpoint inhibitor immunotherapy are possible even in patients unselected for biomarker status. The multicenter open-label phase II CA209-538 clinical trial of nivolumab (3 mg/kg) and ipilimumab (1 mg/kg) every three weeks for four doses followed by nivolumab 3 mg/kg every two weeks for advanced rare cancers included 39 patients with biliary tract cancers, of whom 26 had cholangiocarcinomas, 16 intrahepatic, 10 extrahepatic, and 13 had gallbladder cancer [105]. Among those with cholangiocarcinoma, there were five objective responses, all in the subgroup with intrahepatic tumors, with all responding patients being treated in the second-line setting, and none had microsatellite-unstable tumors. PD-L1 overexpression status was not addressed. Two other patients with intrahepatic cholangiocarcinoma had stable disease for a disease control rate of 44 percent in this group. Duration of response ranged from 3 to 14.8 months. None of the ten patients with extrahepatic cholangiocarcinoma had an objective response, but one patient had stable disease as the best response. In the entire group of 39 patients, approximately one-half experienced an immune-related adverse event of any grade, and ≥grade 3 toxic effects occurred in six (15 percent).

Until further information is available regarding the relevance of additional biomarkers on response rates to combined immunotherapy, based upon this study, a trial of nivolumab plus ipilimumab is reasonable in the setting of second-line therapy for patients with intrahepatic cholangiocarcinoma who do not have dMMR, high levels of TMB, or PD-L1 overexpression, and who did not receive frontline durvalumab. (See 'Gemcitabine plus cisplatin and durvalumab' above.)

FGFR inhibitors for FGFR fusion-positive tumors — Fibroblast growth factor receptor 2 (FGFR2) gene alterations are involved in the pathogenesis of cholangiocarcinoma, and approximately 9 to 16 percent of patients with cholangiocarcinoma (15 to 20 percent of intrahepatic tumors) harbor FGFR2 alterations [90,106-109].

●Pemigatinib – Pemigatinib is a selective, oral inhibitor of FGFR 1, 2, and 3. Activity of this drug in previously treated advanced cholangiocarcinoma was addressed in the open-label single arm FIGHT-202 trial; there were three cohorts: patients with FGFR2 fusions or rearrangements (n = 107), patients with other FGF/FGFR alterations (n = 20), or patients with no FGF/FGFR alterations (n = 18) [90]. All patients initiated therapy at 13.5 mg once daily for 14 days, followed by seven days off. The primary end point was objective response rate. At a median follow-up of 17.8 months, 38 (36 percent) patients with FGFR2 fusions or rearrangements had an objective response including three complete responses. Disease control (objective response or stable disease) was achieved by 88 (80 percent). The median duration of response was 7.5 months (95% CI 5.7-14.5). No patients with other FGF/FGFR alterations or who lacked FGF/FGFR alterations achieved a response, although stable disease was seen in eight of the 20 patients with other FGF/FGFR alterations, and in four (22 percent) of those with no identifiable FGF/FGFR alterations.

In the entire cohort, treatment was well tolerated overall, with the most frequent adverse event being hyperphosphatemia (60 percent, 12 percent grade 3 or worse (table 11)). Hyperphosphatemia occurred early after treatment initiation (median time to onset 15 days [95% CI 8-47]) and was managed with a low-phosphate diet, concomitant phosphate binders, diuretics, dose reduction, and/or dose interruption. Other grade 3 or worse toxicities included arthralgias, stomatitis, hyponatremia, hypophosphatemia, abdominal pain, and fatigue, all in fewer than 7 percent of patients. Serous retinal detachment due to subretinal fluid accumulation occurred in 6 (4 percent) of 146 patients; all events were grade 1 or 2, except for one grade 3 event. Dry eye occurred in 21 percent overall but was severe in only 1 percent. (See "Ocular side effects of systemically administered chemotherapy", section on 'Fibroblast growth factor receptor (FGFR) inhibitors'.)

Largely based on these data, pemigatinib has been approved by the US FDA for adult patients with previously treated, unresectable, locally advanced or metastatic cholangiocarcinoma with a FGFR2 fusion or other rearrangement as detected by an FDA-approved test [110]. The approved dose is 13.5 mg once daily for 14 consecutive days followed by seven days off [111].

On the basis of these encouraging results, an international, phase III, randomized, active-controlled trial is currently recruiting patients to compare pemigatinib with gemcitabine plus cisplatin chemotherapy as initial therapy for unresectable or metastatic cholangiocarcinoma with FGFR2 rearrangements (FIGHT-302; NCT03656536).

●Infigratinib – Infigratinib is another FGFR1-3 selective oral tyrosine kinase inhibitor. Activity in patients with advanced cholangiocarcinoma, progression on one or more lines of systemic therapy, and FGFR2 gene fusions or rearrangements was shown in a preliminary report of a phase II trial, presented at the 2021 ASCO Gastrointestinal Cancers Symposium [112]. Of the 108 patients enrolled, 83 (77 percent) had FGFR2 gene fusions and the remainder had rearrangements; all patients received the drug at 125 mg orally daily for 21 days of each 28-day cycle. At a median follow-up of 10.6 months, there were 25 objective responses (23 percent), one complete, and the median duration of response was five months (range 0.9 to 19.1 months). Responses were more likely in patients treated in the second-line versus the third- or later-line setting (34 versus 13 percent). The most common treatment-emergent adverse effects were hyperphosphatemia (76 percent), eye disorders (68 percent, including one grade 3 central serous retinopathy), stomatitis (55 percent), and fatigue (40 percent). The most common severe adverse events were stomatitis (15 percent), hyponatremia, and hypophosphatemia (13 percent each). (See "Ocular side effects of systemically administered chemotherapy", section on 'Fibroblast growth factor receptor (FGFR) inhibitors'.)

Largely based upon these data, infigratinib was granted accelerated approval by the US FDA for adult patients with previously treated, unresectable locally advanced or metastatic cholangiocarcinoma with a FGFR2 fusion or other rearrangement as detected by an FDA-approved test [113]. The approved dose of infigratinib is 125 mg orally once daily for 21 consecutive days, followed by 7 days off therapy, in 28-day treatment cycles, until disease progression or unacceptable toxicity.

●Futibatinib – Futibatinib is a highly selective, irreversible FGFR1–4 inhibitor. The approved dose is 20 mg orally, once daily, until disease progression or unacceptable toxicity.

Futibatinib was evaluated in an open-label phase II trial (FOENIX-CCA2) of 103 patients with intrahepatic cholangiocarcinoma harboring FGFR2 fusion/rearrangements with disease progression after one or more prior treatments [114]. At median follow-up of 17 months, the confirmed objective response rate was 42 percent (43 patients) and the median duration of response was 10 months. Median progression-free survival was nine months, with a one-year progression-free survival rate of 40 percent. Median overall survival was 22 months, with a one-year overall survival rate of 72 percent. The most common treatment-related adverse reactions were hyperphosphatemia (85 percent), alopecia (33 percent), dry mouth (30 percent), diarrhea (28 percent), dry skin (27 percent), and fatigue (25 percent).

Based on these data, futibatinib was granted an accelerated approval by the US FDA for treatment of locally advanced/metastatic intrahepatic cholangiocarcinoma with an FGFR2 gene rearrangement or fusion [115].

●Erdafitinib – Erdafitinib is an oral selective pan-FGFR tyrosine kinase inhibitor. The use of erdafitinib remains investigational in patients with FGFR-altered biliary tract tumors.

Erdafitinib was evaluated in a phase II trial (RAGNAR) of 217 patients with metastatic tumor of any histology (except urothelial cancer) with a known FGFR1-4 alteration [116]. Among the subgroup of 31 patients with cholangiocarcinoma, objective responses were seen in 16 patients (52 percent).

Ivosidenib for IDH-mutated cholangiocarcinoma — Mutations in the isocitrate dehydrogenase (IDH) genes are present in up to 25 percent (IDH1) and approximately 3 percent (IDH2) of cholangiocarcinomas; mutations are particularly frequent (up to 20 percent) in intrahepatic cases [80,91,117,118], for which specific inhibitors are available (eg, ivosidenib, enasidenib). As an example, in a combined phase I/II study of ivosidenib in 73 patients with advanced IDH1-mutated cholangiocarcinoma, there were four partial responses, and median progression-free survival was 3.8 months [91]. IDH1 is mutated in up to 25 percent of cholangiocarcinomas, especially intrahepatic [91]; the frequency of IDH2 mutations is lower (3 percent in one study [117]).

Benefit for ivosidenib in patients with IDH1-mutant advanced cholangiocarcinoma was further addressed in the placebo-controlled randomized phase III ClarlDHy trial, which enrolled 187 patients with previously treated, progressive advanced (93 percent metastatic) cholangiocarcinoma. The ivosidenib dose was 500 mg once daily in continuous 28-day cycles. In the latest analysis individuals receiving ivosidenib had modestly but significantly better median progression-free survival and a trend toward better overall survival (10.3 versus 7.5 months, HR 0.79, 95% CI 0.56-1.12) [119]. This result was likely influenced by crossover from the placebo to active treatment group; the crossover-adjusted overall survival in the placebo group (as derived using the rank-preserving structural failure time model) was 5.1 months (HR for death with ivosidenib 0.49, 95% CI 0.34-0.70). Notably, in an earlier analysis, 32 percent of the patients receiving ivosidenib had not progressed at six months, and 22 percent were still progression free at 12 months; no patient in the placebo arm lacked disease progression at the six-month point [120]. Common treatment-related adverse effects included nausea, diarrhea, fatigue, cough, abdominal pain, ascites, anorexia, anemia, and vomiting; most were low grade.

Largely based upon these data, ivosidenib has been approved by the US FDA for treatment of adults with previously treated, locally advanced, or metastatic cholangiocarcinoma with an IDH1 mutation, as detected by an FDA-approved test [121]. While this drug may provide modest benefit for a small subset of patients, it may cost as much as USD $1000 per day, a fact that will likely limit general access.

TRK inhibitor therapy for TRK fusion-positive cancers — Another option for treatment at progression for patients who have neurotrophic tyrosine receptor kinase (NTRK) gene rearrangements is larotrectinib or entrectinib. Genomic translocations that lead to the constitutive activation of receptor tyrosine kinases are rare overall in cholangiocarcinoma (3.6 percent for intrahepatic cholangiocarcinoma in one report) [106,122]. However, the potential efficacy of larotrectinib, a highly selective tropomyosin receptor kinase (TRK) inhibitor, was shown in a combined analysis of 55 patients with various TRK fusion-positive malignancies enrolled in three trials, two of whom had a primary cholangiocarcinoma [123]. In the entire cohort, the overall response rate by independent review was 75 percent, and responses appeared durable, with 86 percent of responders still on treatment or having undergone surgery that was intended to be curative at a median follow-up of 9.4 months. One of the two cholangiocarcinoma tumors had objective tumor shrinkage. Treatment was well tolerated; no responding patients discontinued larotrectinib due to an adverse event. In a safety analysis of 176 patients enrolled across all three trials, the most common adverse reactions (≥20 percent) included elevations in transaminases, fatigue, nausea, vomiting, dizziness, diarrhea, constipation, and cough [124].

In November 2018, larotrectinib was approved by the FDA for use in adults and children with solid tumors with an NTRK gene fusion and without a known acquired resistance mutation that are either metastatic or where surgical resection is likely to result in severe morbidity who have no satisfactory alternative treatments or whose cancer has progressed following treatment. A similar drug, entrectinib, was approved in Japan for treatment of 10 tumor types with an NTRK gene fusion (including cholangiocarcinoma), and this drug was approved by the FDA for cancers with an NTRK gene fusion in August 2019. (See "TRK fusion-positive cancers and TRK inhibitor therapy", section on 'Treatment with TRK inhibitors'.)

BRAF V600E-mutated cancers — Specific mutations in the BRAF gene (BRAF V600E) are reported in approximately 5 percent of biliary tract cancers, especially intrahepatic cholangiocarcinomas [125,126]. Benefit from the combination of the BRAF inhibitor dabrafenib plus trametinib (which reversibly and selectively inhibits mitogen-activated extracellular kinase [MEK], a downstream effector of BRAF) was shown in the phase II open-label ROAR trial (47 percent objective response rate by independent assessment among 43 evaluable patients; and additional 14 percent had stable disease) [127]. The median progression-free survival was nine months. Most of these patients had received two or more lines of prior therapy. Benefit for combined dabrafenib and trametinib was also shown in the NCI-MATCH Trial Subprotocol H, in which three of four advanced cholangiocarcinomas had a partial response, one of which was ongoing at 29 months [128].

The combination of dabrafenib and trametinib has accelerated approval from the FDA for the treatment of adult and pediatric patients one year of age and older with unresectable or metastatic solid tumors harboring mutations in BRAF V600E following prior treatment and with no satisfactory alternative treatment options [129,130].

RET fusion-positive tumors — RET fusions are rare in cholangiocarcinoma [131]. However, in September 2022, the US FDA granted a tissue-agnostic, accelerated approval to the RET tyrosine kinase inhibitor selpercatinib for adult patients with locally advanced or metastatic solid tumors, including cholangiocarcinomas, with a RET gene fusion and disease progression on or following prior systemic treatment who have no satisfactory alternative treatment options. The phase I/II basket trial (LIBRETTO-001) of selpercatinib for RET fusion positive solid tumors included only one patient with cholangiocarcinoma, but there was a partial response [132].

HER2-positive tumors — Regimens that target human epidermal growth factor receptor 2 (HER2) are used as later-line therapy for HER2-positive metastatic cholangiocarcinoma. Amplification of the HER2 oncogene or overexpression of the HER2 protein is seen among approximately 5 to 20 percent of cholangiocarcinomas [80,100,133-135]. HER2 molecular alterations are particularly enriched in extrahepatic cholangiocarcinoma.

Selection of therapy — For patients with advanced or metastatic HER2-positive cholangiocarcinoma who progress on gemcitabine, platinum, and/or fluorouracil-based chemotherapy and have no prior exposure to HER2-targeted agents, we suggest either trastuzumab plus pertuzumab or trastuzumab plus tucatinib rather than other systemic agents, as these regimens have good objective response rates with an acceptable toxicity profile. Either of these agents are appropriate as they have not been directly compared in randomized trials. (See 'Trastuzumab plus pertuzumab' below and 'Trastuzumab plus tucatinib' below.)

Fam-trastuzumab deruxtecan is an alternative option, but patients should be cautioned about the risk of interstitial lung disease. (See 'Fam-trastuzumab deruxtecan' below.)

Trastuzumab plus pertuzumab — Trastuzumab plus pertuzumab is a preferred option for later-line therapy in HER2-positive metastatic cholangiocarcinoma.

In an open-label phase II basket study (MyPathway), 39 patients with previously treated, HER2 amplified/overexpressed and RAS wildtype biliary tract cancers were treated with trastuzumab plus pertuzumab. At median follow-up of eight months, objective responses were seen in 9 patients (23 percent) [136]. Median PFS and OS were 4 and 11 months, respectively. The combination was well-tolerated (grade ≥3 toxicity rate and serious treatment-emergent adverse event rate of 8 and 25 percent, respectively).

Trastuzumab plus tucatinib — Trastuzumab plus tucatinib, an oral tyrosine kinase inhibitor highly selective for HER2, is a preferred option for later-line therapy in HER2-positive metastatic cholangiocarcinoma.

Trastuzumab plus tucatinib was evaluated in a phase II basket study (SGNTUC-019) of 30 patients with previously treated HER2 overexpressing or amplified metastatic biliary tract cancer and no prior exposure to HER2 targeted therapy [137]. At median follow-up of 11 months, objective responses were seen in 14 patients (47 percent). Median PFS and OS were 6 and 16 months, respectively. Grade ≥3 toxicity rate was 60 percent, including cholangitis, decreased appetite, and nausea (10 percent each), which were generally not related to therapy.

Fam-trastuzumab deruxtecan — Fam-trastuzumab deruxtecan, an HER2-directed antibody drug conjugate, is an option for treatment-refractory HER2-positive cholangiocarcinoma. However, patients should be cautioned about the risk of interstitial lung disease with this agent.

The efficacy of fam-trastuzumab deruxtecan in treatment-refractory cholangiocarcinoma has been demonstrated in early phase clinical trials [138,139]. In a phase II trial (HERB), fam-trastuzumab deruxtecan was evaluated in a subgroup of 24 patients with HER2-positive biliary tract cancers who progressed on prior gemcitabine-based therapy. In preliminary results, objective responses were seen in 8 of 22 patients (36 percent). Median PFS and OS were 4 and 7 months, respectively [138]. Some clinical efficacy was also seen in the subgroup of eight patients with HER2 low tumors. However, the rate of any grade treatment-related interstitial lung disease in the entire cohort was 25 percent. In another phase II trial (DESTINY-PanTumor02), similar results were seen among the subgroup of patients with biliary tract tumors [139].

Other agents

●Zanidatamab – Zanidatamab, a bispecific antibody targeting two distinct HER2 epitopes, is investigational for the treatment of cholangiocarcinoma. Zanidatamab was evaluated in an open-label phase II trial (HERIZON-BTC-01) of 80 patients with advanced or metastatic HER2-positive biliary tract cancers who progressed on gemcitabine-based therapy and had no prior exposure to HER2-targeted therapy [140]. At median follow-up of 12 months, the objective response rate was 41 percent. Median PFS and OS were 6 and not reached. No responses to zanidatamab were seen among the seven patients with HER2-negative or low tumors.

●Lapatinib – Data suggest minimal efficacy with single-agent lapatinib in previously treated HER2-positive biliary tumors [141,142].

KRAS G12C mutant tumors — Patients with treatment-refractory metastatic cholangiocarcinoma that harbors a KRAS G12C mutation should be encouraged to enroll in clinical trials, where available.

KRAS G12C mutations occur in approximately one percent of patients with biliary tract tumors [143,144]. Studies are ongoing to evaluate agents that target KRAS G12C, such as adagrasib. The use of adagrasib remains investigational in KRAS G12C mutant cholangiocarcinoma.

●Adagrasib – Adagrasib was evaluated in an open-label phase II trial (KRYSTAL-1) of 64 patients with KRAS G12C mutated, treatment-refractory solid tumors of various histologies, excluding non-small cell lung cancer and colorectal cancer [145]. Among the subgroup of 12 patients with biliary tract tumors, objective responses were seen in five patients (42 percent). The objective response rate among those with cholangiocarcinomas was 50 percent (four of eight patients). Median PFS and OS for those with biliary tract tumors was 9 and 15 months, respectively.

What is the role of EGFR inhibitors? — There is no established role for EGFR inhibitors in the treatment of metastatic cholangiocarcinoma. The addition of cetuximab or panitumumab to chemotherapy failed to improve overall survival in randomized trials [146,147]. Early phase clinical trials have demonstrated clinical efficacy for erlotinib as a single agent or in combination with bevacizumab, [71,148], but further data are necessary prior to the routine use of this agent in metastatic cholangiocarcinoma.

PROGNOSTIC FACTORS — Patients with advanced biliary tract cancer are a heterogeneous group, comprising both locally advanced and metastatic disease, as well as a variety of primary disease sites (intrahepatic bile ducts, extrahepatic bile ducts, gallbladder, and ampulla). This heterogeneity complicates assessment of treatment efficacy.

The prognostic influence of these and other factors in patients with advanced biliary tract cancer can be illustrated by a Korean series of 213 patients with advanced biliary tract cancer who were either enrolled in one of two prospective phase II studies or included in a retrospective cohort study of patients receiving initial chemotherapy [149]. In a multivariate analysis, metastatic disease, an intrahepatic primary site, the presence of liver metastases, a poor Eastern Cooperative Oncology Group (ECOG) performance status (table 1), and an elevated level of serum alkaline phosphatase were significant predictors of overall survival. Using these five variables, the authors developed a prognostic index to stratify patients into low, intermediate, and high-risk groups with different median (11.5, 7.3, and 3.6 months, respectively) and one-year survivals (48, 26, and 4 percent, respectively). Although these results are not surprising, they demonstrate the variability of the clinical presentation and outcomes among patients with advanced biliary tract cancer. The broad spectrum of clinical courses and the range of prognostic factors must be kept in mind when assessing the results of investigational studies of new treatments.

Prognostic nomograms based on other factors have also been developed [150]. An important point is that these are prognostic and not predictive models, and they cannot be used to predict which patients are more versus less likely to respond to systemic chemotherapy.

DRUG SHORTAGES — There may be any number of cancer therapies in short supply at various times. Guidance in the setting of drug shortages has been provided by the American Society of Clinical Oncology (table 4).

SUMMARY AND RECOMMENDATIONS

●Initial therapy – The approach to systemic therapy for advanced or metastatic cholangiocarcinoma is evolving. We encourage enrollment in clinical trials, where available. For patients who decline or are ineligible for clinical trials, our approach to initial therapy is based on patient performance status and bilirubin levels.

•Good performance status and no hyperbilirubinemia – For most patients with an Eastern Cooperative Oncology Group (ECOG) performance status (PS) less than two (table 1) and no hyperbilirubinemia (ie, total bilirubin less than 2.5 times the upper limit of normal), we suggest the addition of immunotherapy (either durvalumab or pembrolizumab) to gemcitabine plus cisplatin (table 2) as initial treatment (Grade 2B), as these chemoimmunotherapy regimens improved overall survival and demonstrated durable responses in randomized trials. (See 'Good performance status and no hyperbilirubinemia' above.)

-Although either chemoimmunotherapy regimen is appropriate, gemcitabine plus cisplatin and durvalumab has the advantage of a simpler maintenance therapy with single-agent durvalumab. (See 'Gemcitabine plus cisplatin and durvalumab' above and 'Gemcitabine plus cisplatin and pembrolizumab' above.)

-For patients who decline chemoimmunotherapy or are anticipated to not tolerate its toxicities, gemcitabine plus cisplatin (table 2) is an appropriate alternative. (See 'Gemcitabine plus cisplatin' above.)

-Other options for initial therapy include gemcitabine plus oxaliplatin, gemcitabine plus capecitabine, gemcitabine plus nabpaclitaxel, and gemcitabine plus S-1 (where available). (See 'Other regimens' above.)

•Good performance status, persistent hyperbilirubinemia – For patients with a good PS who have hyperbilirubinemia despite stenting, we prefer a non-gemcitabine-based regimen, such as LV-modulated FU (table 6) or a fluoropyrimidine plus oxaliplatin. (See 'Good performance status and persistent biliary obstruction' above.)

•Borderline performance status or extensive comorbidity – Reasonable initial therapy options for patients with a borderline PS or extensive comorbidity include LV-modulated FU (table 6), capecitabine monotherapy, gemcitabine monotherapy, or S-1 alone (where available). (See 'Borderline performance status' above.)

●Later lines of therapy

•Prior gemcitabine plus cisplatin – For most patients who have disease progression while receiving gemcitabine plus cisplatin and who retain an adequate performance status and lack potentially actionable molecular targets, we suggest treatment with FOLFOX (table 8) (Grade 2C). Other alternatives include capecitabine plus oxaliplatin (CAPOX (table 12)), liposomal irinotecan plus LV-modulated FU (table 9), a fluoropyrimidine alone, or GEMOX with or without bevacizumab. (See 'Gemcitabine plus oxaliplatin' above and 'Patients initially treated with gemcitabine plus cisplatin' above.)

•Prior gemcitabine plus oxaliplatin – For patients failing GEMOX, appropriate choices for a second-line regimen include capecitabine plus cisplatin or gemcitabine, or short-term infusional FU plus LV and irinotecan (FOLFIRI) with or without bevacizumab (table 13). (See 'Cytotoxic chemotherapy' above and "Treatment protocols for hepatobiliary cancer".)

•Molecularly targeted therapy – Most patients with advanced cholangiocarcinomas may harbor a molecular alteration for which targeted treatment is available. These include (see 'Molecularly targeted therapy' above):

-dMMR, PD-L1 overexpression, or high TMB – Immunotherapy for patients with deficient DNA mismatch repair (dMMR), overexpression of programmed cell death ligand 1 (PD-L1), or high levels of tumor mutational burden (TMB) who did not receive durvalumab or pembrolizumab as part of initial therapy. (See 'Biomarker-selected patients' above.)

-Fibroblast growth factor receptor 2 (FGFR2) alteration – Pemigatinib, infigratinib, or futibatinib. (See 'FGFR inhibitors for FGFR fusion-positive tumors' above.)

-Isocitrate dehydrogenase (IDH) mutation – Ivosidenib. (See 'Ivosidenib for IDH-mutated cholangiocarcinoma' above.)

-TRK fusion-positive – A tropomyosin receptor kinase (TRK) inhibitor. (See 'TRK inhibitor therapy for TRK fusion-positive cancers' above.)

-RET fusion-positive – Selpercatinib.

-BRAF V600E mutation – The combination of a BRAF and mitogen-activated extracellular kinase (MEK) inhibitor for those with BRAF V600E mutations. (See 'BRAF V600E-mutated cancers' above.)

-HER2-positive disease – For patients with advanced or metastatic human epidermal growth factor receptor 2 (HER2)-positive cholangiocarcinoma who progress on gemcitabine, platinum, and/or fluorouracil-based chemotherapy and have no prior exposure to HER2-targeted agents, we suggest either trastuzumab plus pertuzumab or trastuzumab plus tucatinib rather than other systemic agents (Grade 2C). Fam-trastuzumab deruxtecan is an alternative option, but patients should be cautioned about the risk of interstitial lung disease. (See 'HER2-positive tumors' above.)

-No actionable molecular alteration – For patients without a molecular alteration, or those with molecular alterations for which a targeted treatment is not available, clinical trial enrollment is encouraged, where available. (See 'Next-generation sequencing to identify actionable molecular abnormalities' above.)

•Immunotherapy – For patients with intrahepatic cholangiocarcinoma who do not have dMMR, high levels of TMB, or PD-L1 overexpression, and who did not receive initial treatment with durvalumab or pembrolizumab, a trial of nivolumab plus ipilimumab is reasonable for second-line therapy or beyond. (See 'Biomarker-unselected patients' above.)