Skip to content
Bending the Curve Trial by Trial
?

Bile ducts and gallbladder (digestive system, hepatobiliary)

Biliary tract cancer

How surgery, chemotherapy, immunotherapy and genetically matched drugs have been studied in cancers of the bile ducts and gallbladder.

Gallbladder cancer · 5-year relative survival · United States (SEER Cancer Statistics Review 1975-2017, Table 1.4; survival from the SEER 18 registry areas)

19.2%

Diagnosed 2010-2016. Read these with the caveats under “Then and now”.Source

24 milestones, by decade

  • In ABC-02, 410 people with locally advanced or metastatic cholangiocarcinoma, gallbladder cancer or ampullary cancer were randomised to cisplatin plus gemcitabine or to gemcitabine alone. Median overall survival was 11.7 months with the combination and 8.1 months with gemcitabine (hazard ratio 0.64; 95% CI 0.52 to 0.80), figures from the abstract of the 2010 New England Journal of Medicine paper.
  • SEER's Cancer Statistics Review 1975-2017 (an archived edition released in April 2020) lists gallbladder cancer at 1.2 new cases per 100,000 people a year in 2013 to 2017 (0.9 in men and 1.5 in women) and the 'other biliary' site group at 1.9 (2.3 in men and 1.5 in women). These are incidence rates age-adjusted to the 2000 US standard population, from the SEER 21 registry areas. The same table has a separate 'Liver & intrahepatic bile duct' row, which is not part of these two figures. SEER's Cancer Stat Facts index has no page for the gallbladder or the biliary tract.
  • FDA approval letters that provide for a biliary tract cancer or cholangiocarcinoma indication include those dated 17 April 2020 (pemigatinib), 25 August 2021 (ivosidenib), 2 September 2022 (durvalumab with gemcitabine and cisplatin), 30 September 2022 (futibatinib, an effective date restated in a corrected letter), 31 October 2023 (pembrolizumab with gemcitabine and cisplatin), 20 November 2024 (zanidatamab, also restated in a corrected letter), 8 May 2026 (zenocutuzumab) and 23 September 2026 (lirafugratinib). The FDA announced on 16 May 2024 the final withdrawal of the accelerated approval it had granted to infigratinib on 28 May 2021.

Evidence current to October 2026 · Checked against primary sources to the extent possible

A guided tour of 12 milestones. You can pause or step through it yourself.

Filter milestonesshowing all 24
Treatment type
Setting

Chemotherapy

  1. 1965 · Understanding · UnderstandingAdenocarcinoma of the hepatic duct at its bifurcation (1965)In 1965 Gerald Klatskin published a paper titled 'Adenocarcinoma of the hepatic duct at its bifurcation within the porta hepatis', calling it an unusual tumour with distinctive clinical and pathological features. Only the title is available in PubMed. A 2020 biography says the tumour was named after him.Klatskin G, Am J Med 1965. The title describes adenocarcinoma of the hepatic duct at its bifurcation within the porta hepatis as an unusual tumour with distinctive clinical and pathological features. Only the title could be read: the PubMed record has no abstract, so no cohort size, findings or outcomes are given. A 2020 biography states that hilar cholangiocarcinoma is eponymously called Klatskin's tumour.

Molecular origins

  1. 1994 · Understanding · UnderstandingLiver flukes (Opisthorchis, Clonorchis) and cholangiocarcinomaIn 1994 the International Agency for Research on Cancer judged infection with the liver fluke Opisthorchis viverrini to be carcinogenic to humans, and with Clonorchis sinensis probably so; the human data it reviewed concerned cholangiocarcinoma and liver cancer. That year a survey of 12,311 adults in northeast Thailand reported sharply higher odds of cholangiocarcinoma in the most heavily infected group.IARC Monographs (1994): infection with Opisthorchis viverrini is carcinogenic to humans (Group 1); Clonorchis sinensis probably carcinogenic (Group 2A). Haswell-Elkins 1994, cross-sectional survey in northeast Thailand: stool examination of 12,311 adults in 85 villages; ultrasound in 1,807 found 15 preclinical cholangiocarcinomas; adjusted prevalence odds ratio 14.1 (p < 0.05) in the most heavily infected group versus uninfected. Later papers state IARC classified C. sinensis as Group 1 in 2009.
  2. 1996 · Understanding · UnderstandingCholangiocarcinoma classed as intrahepatic, perihilar or distal (1996)In 1996 surgeons at Johns Hopkins proposed sorting cholangiocarcinoma, cancer of the bile ducts, into three groups by where it arises: inside the liver, around the liver hilum, or in the distal duct. In their review of 294 patients explored surgically, two thirds of tumours were perihilar, and resection was reported to improve survival at each site.Nakeeb 1996 (Ann Surg): single-institution review of 294 patients with histologically confirmed cholangiocarcinoma explored surgically at Johns Hopkins over 23 years. Sites: 18 intrahepatic (6%), 196 perihilar (67%), 80 distal (27%). Resectability rose with more distal location (50%, 56%, 91%). In resected tumours, 5-year survival 44%, 11%, 28% and median survival 26, 19, 22 months. Postoperative radiation did not improve survival.

Targeted therapy

  1. 1998 · Understanding · UnderstandingPrimary sclerosing cholangitis: cancer risk and surveillanceA 2025 meta-analysis calls primary sclerosing cholangitis, a chronic bile duct disease, a known risk factor for cancers of the liver and biliary tract; across 51 studies it pooled about nine cholangiocarcinomas per 1,000 person-years. In a Swedish cohort of 512, yearly MRI and CA 19-9 surveillance did not find cancers early enough to support long-term survival.Souza 2025 meta-analysis (51 eligible cohort studies, 26,482 patients): pooled cholangiocarcinoma incidence 9.31 per 1,000 person-years (95% CI 6.84–12.67). Villard 2023, prospective Swedish cohort of 512 patients with yearly MRI/MRCP and CA 19-9: 11 cholangiocarcinomas (2%), median survival 13 months; the authors judged it ineffective for early detection. Barner-Rasmussen 2025, three cohorts with different strategies (1,629 patients): composite endpoint (malignancy, transplantation or liver-related death) cumulative incidence 14.1% with scheduled ERCP, the lowest, and 35.0% with on-demand ERCP.
  2. 2001 · Understanding · UnderstandingPreoperative staging for hilar cholangiocarcinoma, 225 patients (2001)In 2001 a Memorial Sloan Kettering team reported 225 people with hilar cholangiocarcinoma and proposed staging tumours before surgery by how far they spread along the bile ducts, whether the portal vein is involved and whether part of the liver has shrunk. The stage predicted whether the tumour could be removed. An earlier version appeared in 1998.Jarnagin 2001 (Ann Surg), Memorial Sloan Kettering: 225 patients with hilar cholangiocarcinoma, 1991–2000, staged on imaging by ductal involvement, portal vein compromise and lobar atrophy. 65 unresectable; 160 explored with curative intent; 80 resected (62 with hepatic resection, 62 with negative margins). After resection, complications 64%, death 10%. In 219 stageable patients the scheme predicted resectability and negative-margin resection and correlated with metastatic disease and survival.
  3. 2005 · Treatment · Cell therapy & transplantNeoadjuvant chemoradiation then transplant for perihilar diseaseAt the Mayo Clinic, people with early perihilar bile duct cancer that was unresectable, or arose with primary sclerosing cholangitis, were offered radiotherapy, chemotherapy and, if staging allowed, a liver transplant. Of 71 who started, 38 were transplanted: about four in five of those were alive at five years, and about three in five of all 71. A selected series.Mayo Clinic series (Rea 2005), non-randomised: 71 patients with hilar cholangiocarcinoma, unresectable or arising with primary sclerosing cholangitis, entered a protocol of neoadjuvant radiotherapy with chemosensitisation, operative staging and liver transplantation; 38 were transplanted. Survival after transplantation was 92%, 82% and 82% at 1, 3 and 5 years; counted from enrolment in all 71, 79%, 61% and 58%.
  4. 2006 · Understanding · UnderstandingGallbladder cancer: gallstones and a wide geographyA 2006 review combining cancer-registry rates with published studies found gallbladder cancer rare overall but very frequent in some places: the highest rates it reported were among women in Delhi, South Karachi and Quito. A history of gallstones was the strongest risk factor, with a pooled relative risk of about five.Randi 2006 (Int J Cancer): review combining cancer-registry rates with relative risks abstracted from selected studies. Highest reported age-adjusted incidence in women: Delhi 21.5, South Karachi 13.8, Quito 12.9 per 100,000. Female-to-male ratio generally about 3 (range 1 to over 5). Pooled RR for gallstone history 4.9 (95% CI 3.3–7.4); Salmonella typhi/paratyphi 4.8 (1.4–17.3); Helicobacter bilis/pylori 4.3 (2.1–8.8). Observational data.
  5. 2010 · Treatment · ChemotherapyABC-02: cisplatin plus gemcitabine versus gemcitabine aloneIn 410 people with locally advanced or metastatic biliary tract cancer, gemcitabine with cisplatin was compared with gemcitabine alone. Median survival was about 12 months with the combination and about 8 months with gemcitabine alone. Side effects were similar, apart from more low neutrophil counts with the combination.ABC-02 (phase 3, N=410; cholangiocarcinoma, gallbladder or ampullary cancer): cisplatin-gemcitabine (n=204) vs gemcitabine (n=206). Median OS 11.7 vs 8.1 months, HR 0.64 (95% CI 0.52-0.80), P<0.001; median PFS 8.0 vs 5.0 months (P<0.001). BT22 (Japan, randomised phase 2, 83 analysed): median OS 11.2 vs 7.7 months, HR 0.69 (95% CI 0.42-1.13), P=0.139.

Immunotherapy

  1. 2011 · Understanding · UnderstandingIDH1/2 mutations, FGFR2 fusions and etiological subtypesFour genomic studies from 2011 to 2017 described molecular groups within bile duct cancers. One found IDH1 or IDH2 mutations in nearly a quarter of intrahepatic cholangiocarcinomas in its samples, and none in extrahepatic or gallbladder tumours. Another found FGFR2 fusions in about one in seven intrahepatic tumours. A 489-tumour study linked molecular subtypes to liver-fluke infection.Borger 2012: IDH1/IDH2 mutations in 9 of 40 intrahepatic cholangiocarcinomas (23%), none in 22 extrahepatic or 25 gallbladder carcinomas. Arai 2014: FGFR2 fusions in 9 of 66 intrahepatic tumours (13.6%). Nakamura 2015: 260 biliary tract cancers, nearly 40% with targetable alterations. Jusakul 2017: 489 cholangiocarcinomas, 10 countries; fluke-positive tumours enriched for ERBB2 amplification and TP53 mutation, fluke-negative clusters included IDH1/2 and BAP1 mutations and FGFR/PRKA-related rearrangements.
  2. 2018 · Reversal · ChemotherapyBCAT, PRODIGE 12: adjuvant gemcitabine trials, no significant benefitTwo randomised trials tested gemcitabine-based chemotherapy after surgery for bile duct and related cancers: gemcitabine alone in one, gemcitabine with oxaliplatin in the other. Neither found a significant survival difference compared with observation or surveillance, and severe side effects were more frequent with the combination.BCAT (Ebata 2018): 225 patients with resected bile duct cancer, gemcitabine (117) vs observation (108): median OS 62.3 vs 63.8 months, HR 1.01 (95% CI 0.70-1.45), p=0.964. PRODIGE 12 (Edeline 2019): 196 patients, 12 cycles of gemcitabine plus oxaliplatin vs surveillance: median RFS 30.4 vs 18.5 months, HR 0.88 (0.62-1.25), p=0.48; OS 75.8 vs 50.8 months, HR 1.08 (0.70-1.66).
  3. 2018 · Reversal · SurgeryEndoscopic versus percutaneous drainage before perihilar resectionA Dutch randomised trial compared two ways of draining a blocked bile duct before liver surgery for perihilar bile duct cancer. It closed early because more people died in the percutaneous drainage group: eleven of 27 against three of 27. Severe drainage complications were similar. The trial was small, and the authors urged caution.Coelen 2018: multicentre randomised trial at four Dutch academic centres; 54 patients with potentially resectable perihilar cholangiocarcinoma randomised 1:1 to endoscopic (27) or percutaneous transhepatic (27) biliary drainage. Closed prematurely for higher mortality with percutaneous drainage: 11 (41%) vs 3 (11%) with endoscopic, RR 3.67 (95% CI 1.15-11.69), p=0.03. Primary outcome, severe preoperative drainage-related complications (percutaneous vs endoscopic): 63% vs 67%, RR 0.94 (95% CI 0.64-1.40).
  4. 2018 · Reversal · SurgeryPhotodynamic therapy: small positive trials, then PHOTOSTENT-02Two small randomised trials in the 2000s reported longer survival when photodynamic therapy, a light-activated treatment inside the bile duct, was added to a stent. A larger randomised trial of 92 people later reported shorter survival with it: about six months against about ten. Its authors concluded it cannot be recommended outside clinical trials.PHOTOSTENT-02 (Pereira 2018): open-label randomised phase 3, 92 patients with locally advanced or metastatic biliary tract cancer and adequate drainage, porfimer sodium photodynamic therapy plus stenting vs stenting alone (46 per group); recruitment stopped on data-monitoring advice. Median OS 6.2 vs 9.8 months (HR 1.56, 95% CI 1.00-2.43, p=0.048); PFS 3.4 vs 4.3 months (HR 1.43, 0.93-2.18). Earlier randomised trials (Ortner 2003; Zoepf 2005, 32 patients) reported longer survival.
  5. 2019 · Treatment · ChemotherapyBILCAP: adjuvant capecitabine after resection of biliary tract cancerBILCAP randomly assigned 447 people in the UK, after surgery for bile duct or gallbladder cancer, to capecitabine tablets or to observation. In the main analysis the survival difference was not statistically significant: just over four years against three. Two other prespecified analyses pointed towards a benefit, and the authors later wrote that capecitabine has become the standard of care.BILCAP (Primrose 2019): phase 3, 447 patients after resection of cholangiocarcinoma or muscle-invasive gallbladder cancer, capecitabine (eight cycles) vs observation. ITT median OS 51.1 vs 36.4 months, adjusted HR 0.81 (95% CI 0.63-1.04), p=0.097: primary endpoint not met. Sensitivity analysis HR 0.71 (0.55-0.92); per-protocol HR 0.75 (0.58-0.97). Long-term follow-up (Bridgewater 2022): ITT HR 0.84 (0.67-1.06).

Precision & engineered

  1. 2020 · Treatment · Targeted⚑ needs reviewFGFR2 inhibitors: pemigatinib (FIGHT-202) and futibatinibSome bile duct cancers carry a rearrangement of the FGFR2 gene. In single-arm studies of people whose cancer had already been treated, the FGFR2 inhibitors pemigatinib and futibatinib shrank tumours in roughly one in three and four in ten people. The FDA gave both drugs accelerated approval; neither study had a comparison group.Single-arm phase 2 studies of previously treated cholangiocarcinoma with FGFR2 fusion or rearrangement; primary endpoint ORR by central review. FIGHT-202, pemigatinib: ORR 35.5% (95% CI 26.5-45.4), 38 of 107; final analysis (108 patients) ORR 37.0%, median OS 17.5 months. FOENIX-CCA2, futibatinib (intrahepatic only): ORR 42% (32-52), 43 of 103; median DOR 9.7 months, PFS 9.0, OS 21.7. No comparator arms.
  2. 2020 · Treatment · Targeted⚑ needs reviewClarIDHy: ivosidenib versus placebo in IDH1-mutant cholangiocarcinomaIn people with IDH1-mutant bile duct cancer that had progressed on chemotherapy, the pill ivosidenib delayed progression compared with placebo (2.7 versus 1.4 months, a small absolute gain). Median survival was nearly three months longer (10.3 versus 7.5 months), but not statistically significant, and most placebo patients crossed over to ivosidenib.ClarIDHy, phase 3, double-blind, 2:1; previously treated IDH1-mutant cholangiocarcinoma. Primary endpoint PFS by independent review (124 vs 61): median 2.7 vs 1.4 months, HR 0.37 (95% CI 0.25-0.54), one-sided P<0.0001. Final OS (187 patients; 43 placebo crossovers): 10.3 vs 7.5 months, HR 0.79 (0.56-1.12), one-sided P=.09. RPSFT-adjusted placebo median 5.1 months, HR 0.49; 95% CI 0.25-1.24 after a 2026 erratum.
  3. 2020 · Treatment · TargetedBRAF V600E, NRG1 fusions and tissue-agnostic options (2020-2026)Mutations in the BRAF gene are reported in about one in twenty biliary tract tumours. In single-arm studies, dabrafenib plus trametinib produced responses in about half of 43 people with BRAF V600E-mutated disease, and zenocutuzumab in about one in three of 19 people with NRG1 gene fusions. Both options are FDA-approved for people whose tumours carry the alteration.ROAR biliary cohort (phase 2 basket, single-arm; dabrafenib plus trametinib; 43 patients with BRAF V600E-mutated disease; interim analysis): ORR 51% (95% CI 36-67) by investigator, 47% (31-62) by independent review. eNRGy cholangiocarcinoma (zenocutuzumab; FDA label, 19 patients with NRG1 fusion): confirmed ORR 36.8% (16.3-61.6) by BICR, response duration 2.8 to 12.9 months. FDA approved zenocutuzumab for this indication on 8 May 2026.
  4. 2021 · Treatment · ChemotherapyABC-06: second-line FOLFOX added to active symptom controlIn 162 people whose biliary tract cancer had progressed after gemcitabine and cisplatin, adding FOLFOX chemotherapy to active symptom control was compared with active symptom control alone. Median survival was about 6 months with FOLFOX and about 5 months without it. After a year, about one in four people were alive with FOLFOX, against about one in ten without.ABC-06 (phase 3, open-label, UK, N=162; progression on cisplatin-gemcitabine, ECOG 0-1): active symptom control plus FOLFOX (n=81) vs active symptom control alone (n=81). Median OS 6.2 (95% CI 5.4-7.6) vs 5.3 months (4.1-5.8); adjusted HR 0.69 (95% CI 0.50-0.97), P=0.031. OS at 6 months 50.6% vs 35.5%; at 12 months 25.9% vs 11.4%. Grade 3-5 adverse events 69% vs 52%.
  5. 2022 · Treatment · ImmunotherapyTOPAZ-1: durvalumab added to gemcitabine and cisplatinIn 685 people with advanced biliary tract cancer, adding the immunotherapy drug durvalumab to gemcitabine and cisplatin was compared with adding a placebo. About one in four people given durvalumab were alive at two years, compared with about one in ten given placebo. The US FDA approved the combination in September 2022.TOPAZ-1 (phase 3, double-blind, N=685; durvalumab n=341, placebo n=344, each with gemcitabine-cisplatin). Primary analysis: OS HR 0.80 (95% CI 0.66-0.97), P=0.021; 24-month OS 24.9% vs 10.4%; PFS HR 0.75 (95% CI 0.63-0.89). Updates: median OS 12.9 vs 11.3 months, HR 0.76 (95% CI 0.64-0.91); at 41.3 months' follow-up HR 0.74 (95% CI 0.63-0.87), 36-month OS 14.6% vs 6.9%.
  6. 2023 · Treatment · ChemotherapyASCOT (JCOG1202): adjuvant S-1 versus observationIn Japan, 440 people were randomly assigned after surgery for biliary tract cancer to the tablet chemotherapy S-1 or to observation. After three years, about 77 in 100 given S-1 were alive against about 68 in 100 under observation. The authors suggested S-1 could be a standard of care for Asian patients, with longer follow-up still needed.ASCOT/JCOG1202: open-label phase 3 at 38 Japanese hospitals; 440 patients after resection of extrahepatic or intrahepatic cholangiocarcinoma, gallbladder or ampullary carcinoma, S-1 (218) vs observation (222). Primary analysis, ITT: 3-year OS 77.1% (95% CI 70.9-82.1) vs 67.6% (61.0-73.3); adjusted HR 0.69 (0.51-0.94), one-sided p=0.0080. 3-year RFS 62.4% vs 50.9%, HR 0.80 (0.61-1.04), two-sided p=0.088.
  7. 2023 · Treatment · Immunotherapy⚑ needs reviewKEYNOTE-966: pembrolizumab added to gemcitabine and cisplatinIn 1,069 people with advanced biliary tract cancer, adding the immunotherapy drug pembrolizumab to gemcitabine and cisplatin was compared with adding a placebo. Median survival was about 13 months with pembrolizumab and about 11 months with placebo. The US FDA approved the combination in October 2023.KEYNOTE-966 (phase 3, double-blind, N=1069; pembrolizumab n=533, placebo n=536, each with gemcitabine-cisplatin). ITT: median OS 12.7 (95% CI 11.5-13.6) vs 10.9 months (9.9-11.6); HR 0.83 (95% CI 0.72-0.95); one-sided P=0.0034 (threshold 0.0200); median follow-up 25.6 months. At 36.6 months' follow-up: HR 0.86 (95% CI 0.75-0.98); 36-month OS 13% vs 11%, reported without P values.
  8. 2023 · Treatment · ImmunotherapyHER2-positive disease: zanidatamab and trastuzumab deruxtecanSome biliary tract cancers carry extra copies of the HER2 gene or high HER2 protein levels. In single-arm studies of previously treated patients, HER2-directed drugs (zanidatamab, trastuzumab deruxtecan) produced responses in roughly one in four to one in two people. The FDA gave zanidatamab accelerated approval in 2024; a randomised confirmation trial is required.Single-arm studies in previously treated HER2-positive biliary tract cancer. HERIZON-BTC-01, zanidatamab: cohort 1 (IHC 2+/3+, 80 patients) confirmed ORR 41.3% (95% CI 30.4-52.8); IHC 3+ (label, 62 patients) ORR 52% (39-65), median DOR 14.9 months. DESTINY-PanTumor02 biliary cohort, trastuzumab deruxtecan: ORR 22.0% (10.6-37.6); central IHC 3+ 56.3% (16 patients). HERB: ORR 36.4% (22 patients).
  9. 2024 · Treatment · ChemotherapyOther first-line regimens: S1815, AMEBICA, MITSUBA, JCOG1113Four randomised trials in advanced biliary tract cancer tested other first-line regimens against gemcitabine and cisplatin. Adding nab-paclitaxel, or using modified FOLFIRINOX, showed no survival advantage; in the largest trial, median survival was about 14 months in both groups. Trials with S-1 reported a survival gain when it was added, and survival no worse than cisplatin's when it replaced cisplatin.Control in all four: gemcitabine-cisplatin. S1815 (phase 3, primary end point OS): plus nab-paclitaxel, median OS 14.0 vs 13.6 months, HR 0.91 (95% CI 0.72-1.14), P=0.41. MITSUBA (phase 3, OS): plus S-1, 13.5 vs 12.6 months, HR 0.791 (90% CI 0.628-0.996), P=0.046. JCOG1113 (non-inferiority, OS): gemcitabine-S-1 instead of cisplatin, 15.1 vs 13.4 months, HR 0.945 (90% CI 0.78-1.15). AMEBICA (phase 2 part, primary end point 6-month PFS rate): modified FOLFIRINOX, 44.6% vs 47.3%.
  10. 2024 · Reversal · TargetedFGFR2 first-line trials end early; infigratinib approval withdrawnTwo randomised trials compared an FGFR2 inhibitor with gemcitabine and cisplatin as first treatment. Both closed early, so neither settled the question: FIGHT-302 reported longer progression-free survival with pemigatinib, while PROOF 301 enrolled only 48 people. In 2024 the FDA withdrew infigratinib's accelerated approval at the sponsor's request.FIGHT-302 (phase 3, open-label, 1:1; pemigatinib 83 vs gemcitabine-cisplatin 84; closed early after a change in standard of care): median PFS 8.3 vs 6.8 months, HR 0.58 (95% CI 0.39-0.87), nominal P=.0078; ORR 47% vs 15%; median OS 24.4 vs 25.0 months, with crossover. PROOF 301 (2:1; infigratinib 29 vs 19; terminated for poor accrual): median PFS by BICR 7.4 vs 8.0 months; no inferential analysis.
  11. 2026 · Treatment · ChemotherapyNeoadjuvant GOLP before resection of intrahepatic cholangiocarcinomaA randomised phase 2-3 trial assigned 178 people with high-risk intrahepatic cholangiocarcinoma to chemotherapy, an immunotherapy drug and lenvatinib before surgery, or to surgery first. At an interim analysis, time free of events was longer with treatment before surgery: 18 months against nearly nine. Two-year survival was higher but did not meet the trial's significance criterion.ZSAB-neoGOLP: randomised phase 2-3, 178 patients with resectable high-risk intrahepatic cholangiocarcinoma; neoadjuvant gemcitabine-oxaliplatin, toripalimab and lenvatinib then resection (88) vs resection alone (90); all received adjuvant capecitabine. Interim analysis, median follow-up 16.9 months: median EFS (primary endpoint) 18.0 vs 8.7 months, P<0.001. 24-month OS 79% vs 61%, HR 0.43 (95% CI 0.23-0.79), P=0.005, above the two-sided alpha of 0.0019.

Then and now

  • 5-year relative survival · United States (SEER Cancer Statistics Review 1975-2017, Table 1.4; survival from the SEER 18 registry areas)

    • 19.2%

      Diagnosed 2010-2016

    People of all races and both sexes with cancer of the gallbladder (the 'Gallbladder' row and 'Total' survival column of Table 1.4, All Races), diagnosed in 2010 to 2016 in the SEER 18 registry areas, with follow-up into 2017. The row states no age range and no number of people. In the same table the male and female figures are 20.6% and 18.6%.

    SEER's own review table, not a Cancer Stat Facts figure (SEER's Cancer Stat Facts has no page for the gallbladder or the biliary tract). The table prints only the site label; SEER's site recode (page read) lists Gallbladder as ICD-O-3 topography C239, and whether this edition used that recode was not confirmed. Relative survival compares survival of people with this cancer with the expected survival of a similar cancer-free group; the table footnote says expected survival is derived from life tables by socio-economic status, geography and race. The source is an archived past version of the Cancer Statistics Review (data submission of November 2019, released April 2020). No earlier-period series was found for this site: the Cancer Statistics Review 1975-2017 has no section for the gallbladder or other biliary sites, and its survival-by-year-of-diagnosis file contains a table for the liver and intrahepatic bile duct only. There is therefore no 'then' value to set beside this one and no change over time is implied. Do not compare it with the American Cancer Society figures (diagnosed 2015 to 2021, a different registry set and method) or with the Cancer Research UK figure for England.

    Source

  • 5-year relative survival · United States (SEER Cancer Statistics Review 1975-2017, Table 1.4; survival from the SEER 18 registry areas)

    • 18.7%

      Diagnosed 2010-2016

    People of all races and both sexes with cancer in the SEER site group 'Other biliary' (the 'Other biliary' row and 'Total' survival column of Table 1.4, All Races), diagnosed in 2010 to 2016 in the SEER 18 registry areas, with follow-up into 2017. The row states no age range and no number of people. In the same table the male and female figures are 19.7% and 17.5%.

    SEER's own review table, not a Cancer Stat Facts figure (SEER's Cancer Stat Facts has no page for the gallbladder or the biliary tract). The table prints only the label 'Other biliary'; SEER's site recode (page read) lists Other Biliary as ICD-O-3 topography C240-C249, separate from the gallbladder (C239) and the intrahepatic bile duct (C221), and whether this edition used that recode was not confirmed. Intrahepatic bile duct cancer belongs to the table's 'Liver & intrahepatic bile duct' row (19.6% survival); SEER's Cancer Stat Facts page for the liver and intrahepatic bile duct says its summary refers to primary liver cancer, so that row is not used here as a biliary series. Relative survival compares survival of people with this cancer with the expected survival of a similar cancer-free group; the table footnote says expected survival is derived from life tables by socio-economic status, geography and race. The source is an archived past version of the Cancer Statistics Review (data submission of November 2019, released April 2020). No earlier-period series was found for this site: the Cancer Statistics Review 1975-2017 has no section for the gallbladder or other biliary sites, and its survival-by-year-of-diagnosis file contains a table for the liver and intrahepatic bile duct only. There is therefore no 'then' value to set beside this one and no change over time is implied. Do not compare it with the American Cancer Society figures (diagnosed 2015 to 2021, a different registry set and method) or with the Cancer Research UK figure for England.

    Source

  • Age-standardised mortality · United States (SEER Cancer Statistics Review 1975-2017, Table 1.4; US mortality files)

    • 0.6per 100,000

      Deaths in 2013-2017

    People of all races and both sexes who died of gallbladder cancer in the United States in 2013 to 2017 (the 'Gallbladder' row and 'Total' US mortality column of Table 1.4, All Races). The table states the rate is per 100,000 and age-adjusted to the 2000 US standard population (19 age groups). In the same table the male and female rates are 0.4 and 0.7 per 100,000.

    SEER's own review table, not a Cancer Stat Facts figure (SEER's Cancer Stat Facts has no page for the gallbladder or the biliary tract). US mortality comes from the National Center for Health Statistics files via the Centers for Disease Control and Prevention, as the table footnote states. The rate is printed to one decimal place, which limits comparisons. The table gives separate mortality rows for 'Gallbladder', 'Other biliary' and 'Liver & intrahepatic bile duct'; which row a death from an intrahepatic bile duct tumour falls in depends on cause-of-death coding, which was not examined here. This is a single value for one period, shown beside the survival figures; it implies no change over time. The same review's Table 1.7 gives annual percent changes for 2008 to 2017 (gallbladder mortality -1.2, marked as significantly different from zero; gallbladder incidence -0.3, not marked); these describe change within that decade and are not an earlier-period value for this record. The review's 68-year death-rate table (Table 1.2) counts gallbladder and biliary tract, not otherwise specified, within its liver and intrahepatic bile duct group, so no long-term mortality series for this site is available there.

    Source

  • Age-standardised mortality · United States (SEER Cancer Statistics Review 1975-2017, Table 1.4; US mortality files)

    • 0.4per 100,000

      Deaths in 2013-2017

    People of all races and both sexes who died of cancer in the SEER site group 'Other biliary' in the United States in 2013 to 2017 (the 'Other biliary' row and 'Total' US mortality column of Table 1.4, All Races). The table states the rate is per 100,000 and age-adjusted to the 2000 US standard population (19 age groups). In the same table the male and female rates are 0.5 and 0.4 per 100,000.

    SEER's own review table, not a Cancer Stat Facts figure (SEER's Cancer Stat Facts has no page for the gallbladder or the biliary tract). Same series notes as the gallbladder mortality record: the table prints only the label 'Other biliary' (SEER's site recode lists C240-C249), the rate is rounded to one decimal place, and this is a single value for one period. By the table's labels, intrahepatic bile duct cancer is not in this row (the table has a separate 'Liver & intrahepatic bile duct' row). The same review's Table 1.7 gives annual percent changes for 2008 to 2017 for this site group (mortality 0.8 and incidence -0.2, neither marked as significantly different from zero); these describe change within that decade and are not an earlier-period value, and the Table 1.2 limitation described in the gallbladder mortality record applies here too.

    Source

  • 5-year relative survival · United States (American Cancer Society)

    • 10%

      Diagnosed 2015-2021, all SEER stages combined

    People diagnosed with intrahepatic bile duct cancer (cancer that starts in the bile ducts inside the liver, as the page describes it) between 2015 and 2021 in the United States, all SEER stages combined, as given in the intrahepatic table of the American Cancer Society page 'Survival Rates for Bile Duct Cancer'. The page states no age range, sex or number of people. Its stage rows for this table read localized 25%, regional 12% and distant 3%.

    Secondary source: the American Cancer Society says the numbers are based on SEER (its reference is SEER*Explorer, accessed 13 June 2025) but does not name the registry set or the site definitions. A relative survival rate compares people with this cancer with the general population. By the page's own list, the figures do not account for age, overall health or how well the cancer responds to treatment, they apply to stage at first diagnosis only, and people diagnosed more recently may have a better outlook because the figures describe people diagnosed and treated at least five years earlier. The page keeps intrahepatic and extrahepatic bile duct cancer in separate tables, and they are different populations from each other, from gallbladder cancer and from the SEER Cancer Statistics Review rows above (different period, registry set and method). No earlier-period series for intrahepatic bile duct cancer alone was available from the sources used, so there is no 'then' value and no change over time is implied.

    Source

  • 5-year relative survival · United States (American Cancer Society)

    • 13%

      Diagnosed 2015-2021, all SEER stages combined

    People diagnosed with extrahepatic bile duct cancer (cancer that starts in the bile ducts outside the liver; the page says this includes perihilar and distal bile duct cancers) between 2015 and 2021 in the United States, all SEER stages combined, as given in the extrahepatic table of the American Cancer Society page 'Survival Rates for Bile Duct Cancer'. The page states no age range, sex or number of people. Its stage rows for this table read localized 19%, regional 20% and distant 2%.

    Secondary source: the American Cancer Society says the numbers are based on SEER (its reference is SEER*Explorer, accessed 13 June 2025) but does not name the registry set or the site definitions. A relative survival rate compares people with this cancer with the general population. By the page's own list, the figures do not account for age, overall health or how well the cancer responds to treatment, they apply to stage at first diagnosis only, and people diagnosed more recently may have a better outlook because the figures describe people diagnosed and treated at least five years earlier. In this table the regional figure (20%) is higher than the localized figure (19%), so the stage groups are not ordered as might be expected and should not be over-interpreted. Extrahepatic and intrahepatic figures are different populations; no earlier-period series was available, so no change over time is implied.

    Source

  • 5-year relative survival · United States (American Cancer Society)

    • 67%

      Diagnosed 2015-2021, localized stage

    Gallbladder cancer, localized stage at diagnosis, for people diagnosed between 2015 and 2021 in the United States, as given on the American Cancer Society page 'Survival Rates for Gallbladder Cancer'. Localized means no sign that the cancer has spread outside the gallbladder, regional means it has spread outside the gallbladder to nearby structures or lymph nodes, and distant means it has spread to distant parts of the body, such as the lungs. The page states no age range, sex or number of people and gives no all-stages figure.

    Secondary source: the American Cancer Society says the numbers come from SEER*Explorer (SEER Incidence Data, November 2024 Submission, 1975-2022, SEER 21 registries excluding Illinois, per the page's reference). A relative survival rate compares people with this cancer with the general population. By the page's own wording, the numbers apply only to the stage when the cancer is first diagnosed, they do not take everything into account (age, overall health and how well the cancer responds to treatment also matter) and people diagnosed now may have a better outlook because the figures describe people diagnosed and treated at least five years earlier. The page prints no all-stages figure, so the three stage records are deliberately worded as different populations and are not a change over time. Stage-specific figures only.

    Source

  • 5-year relative survival · United States (American Cancer Society)

    • 29%

      Diagnosed 2015-2021, regional stage

    Gallbladder cancer, regional stage at diagnosis, for people diagnosed between 2015 and 2021 in the United States, as given on the American Cancer Society page 'Survival Rates for Gallbladder Cancer'. Localized means no sign that the cancer has spread outside the gallbladder, regional means it has spread outside the gallbladder to nearby structures or lymph nodes, and distant means it has spread to distant parts of the body, such as the lungs. The page states no age range, sex or number of people and gives no all-stages figure.

    Secondary source: the American Cancer Society says the numbers come from SEER*Explorer (SEER Incidence Data, November 2024 Submission, 1975-2022, SEER 21 registries excluding Illinois, per the page's reference). A relative survival rate compares people with this cancer with the general population. By the page's own wording, the numbers apply only to the stage when the cancer is first diagnosed, they do not take everything into account (age, overall health and how well the cancer responds to treatment also matter) and people diagnosed now may have a better outlook because the figures describe people diagnosed and treated at least five years earlier. The page prints no all-stages figure, so the three stage records are deliberately worded as different populations and are not a change over time. Stage-specific figures only.

    Source

  • 5-year relative survival · United States (American Cancer Society)

    • 4%

      Diagnosed 2015-2021, distant stage

    Gallbladder cancer, distant stage at diagnosis, for people diagnosed between 2015 and 2021 in the United States, as given on the American Cancer Society page 'Survival Rates for Gallbladder Cancer'. Localized means no sign that the cancer has spread outside the gallbladder, regional means it has spread outside the gallbladder to nearby structures or lymph nodes, and distant means it has spread to distant parts of the body, such as the lungs. The page states no age range, sex or number of people and gives no all-stages figure.

    Secondary source: the American Cancer Society says the numbers come from SEER*Explorer (SEER Incidence Data, November 2024 Submission, 1975-2022, SEER 21 registries excluding Illinois, per the page's reference). A relative survival rate compares people with this cancer with the general population. By the page's own wording, the numbers apply only to the stage when the cancer is first diagnosed, they do not take everything into account (age, overall health and how well the cancer responds to treatment also matter) and people diagnosed now may have a better outlook because the figures describe people diagnosed and treated at least five years earlier. The page prints no all-stages figure, so the three stage records are deliberately worded as different populations and are not a change over time. Stage-specific figures only.

    Source

  • 5-year relative survival · England (Cancer Research UK)

    • 11.8%

      Diagnosed 2016-2020

    People diagnosed with gallbladder cancer in England between 2016 and 2020, as given on the Cancer Research UK health professional page 'Gallbladder cancer survival statistics' (the page states no age range for this headline figure). The page also gives 18.2% for people diagnosed at ages 15 to 54 and 6.7% for ages 75 to 99, and says five-year survival is similar in females and males (no figures in the text read).

    Secondary source: the page words the figure as the proportion who 'survive their disease for five years or more' and elsewhere refers to five-year relative survival in the UK; the text read does not name the registry source or state net survival. Cancer Research UK has no bile duct cancer statistics page at the addresses tried (HTTP 404), so this series covers the gallbladder only. It is a different country, period and method from the US figures and is not comparable with them. No earlier-period series was read, so no change over time is implied.

    Source

Survival figures are affected by earlier detection (lead time) and overdiagnosis, and by changes in who is diagnosed. For screened cancers, read them alongside mortality.

Evidence chains

A lane follows one defined population and setting, where each new trial tests against what was then standard. Compare only within a lane, and read every comparison as illustrative.

Advanced biliary tract cancer, first line: overall survival

Adults with previously untreated locally advanced, unresectable or metastatic biliary tract cancer (cholangiocarcinoma or gallbladder cancer; ABC-02 also enrolled ampullary cancer and TOPAZ-1 also recurrent disease) · Metastatic · endpoint: OS

Results are not comparable across trials.

The control arms differ: gemcitabine alone in ABC-02 and BT22, and gemcitabine with cisplatin in the later trials (with a placebo in TOPAZ-1 and KEYNOTE-966). The experimental arms differ in kind: adding cisplatin (ABC-02, BT22), adding durvalumab or pembrolizumab (TOPAZ-1, KEYNOTE-966), adding nab-paclitaxel (S1815) or S-1 (MITSUBA), and, in JCOG1113, replacing cisplatin with S-1 in a non-inferiority design, so that experimental arm is a substitution tested for survival no worse than the control, not a test of superiority. Populations differ: ABC-02 also enrolled ampullary cancer, TOPAZ-1 also recurrent disease, KEYNOTE-966 required measurable disease and ECOG performance status 0 or 1, and JCOG1113 enrolled people in Japan with ECOG 0 or 1. BT22 is a randomised phase 2 trial whose primary objective was the 1-year survival rate; its arm sizes (41 and 42) are the 83 patients analysed, not the 84 randomised, and the interval of its hazard ratio includes 1. Follow-up and analysis stage differ: ABC-02 reports a median follow-up of 8.2 months; KEYNOTE-966 is drawn from its final analysis (median follow-up 25.6 months, one-sided P value); the TOPAZ-1 medians (12.8 versus 11.5 months) belong to its primary analysis, a prespecified interim analysis, and come from the FDA prescribing information because the paper's abstract gives none, while its later updates are separate results and are not drawn. Hazard-ratio intervals are 95% intervals except in MITSUBA and JCOG1113, which print 90% intervals. S1815, MITSUBA, JCOG1113 and AMEBICA share one milestone dated 2024 by S1815; JCOG1113 was reported in 2019 and MITSUBA before 2024, so the chart shows them later than their own publication. AMEBICA (phase 2 part, overall survival a secondary and unpowered result) is not drawn. Compare only within each trial's own two arms; the lane makes no recommendation and no claim that regimens are equivalent.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

05101520OS, median (months)2010201020222023202420242024ABC-02BT22TOPAZ-1KEYNOTE-966JCOG1113/FUGA-BTKHBO1401-MITSUBASWOG S1815
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
OS, median, across 7 trials in this lane. ABC-02 (2010): experimental 11.7 months vs control 8.1 months; BT22 (2010): experimental 11.2 months vs control 7.7 months; TOPAZ-1 (2022): experimental 12.8 months vs control 11.5 months; KEYNOTE-966 (2023): experimental 12.7 months vs control 10.9 months; JCOG1113/FUGA-BT (2024): experimental 15.1 months vs control 13.4 months; KHBO1401-MITSUBA (2024): experimental 13.5 months vs control 12.6 months; SWOG S1815 (2024): experimental 14 months vs control 13.6 months.
Data table
TrialYearEndpointControl (months)Experimental (months)
ABC-022010OS8.1 months11.7 months
BT222010OS7.7 months11.2 months
TOPAZ-12022OS11.5 months12.8 months
KEYNOTE-9662023OS10.9 months12.7 months
JCOG1113/FUGA-BT2024OS13.4 months15.1 months
KHBO1401-MITSUBA2024OS12.6 months13.5 months
SWOG S18152024OS13.6 months14 months

Resected biliary tract cancer, adjuvant setting: overall survival

Adults after resection of biliary tract cancer, randomised to adjuvant chemotherapy or to observation or surveillance (BCAT: extrahepatic bile duct cancer only) · Adjuvant · endpoint: OS

Results are not comparable across trials.

The trials differ in population, primary endpoint, control and treatment. BCAT enrolled extrahepatic bile duct cancer only (perihilar and distal; intrahepatic tumours and gallbladder cancer were not eligible) and compared gemcitabine with observation. PRODIGE 12 enrolled localised biliary tract cancer after R0 or R1 resection and compared 12 cycles of gemcitabine plus oxaliplatin with surveillance. BILCAP enrolled cholangiocarcinoma or muscle-invasive gallbladder cancer and compared eight cycles of capecitabine with observation. ASCOT enrolled extrahepatic or intrahepatic cholangiocarcinoma, gallbladder carcinoma or ampullary carcinoma at 38 Japanese hospitals and compared four cycles of S-1 with observation. The primary endpoint was overall survival in BCAT, BILCAP and ASCOT, but relapse-free survival and health-related quality of life in PRODIGE 12, where overall survival is a secondary result. In PRODIGE 12 the median overall survival values (75.8 versus 50.8 months) point the other way from its hazard ratio of 1.08 (95% CI 0.70 to 1.66), and the abstract gives no intervals for the medians. BILCAP is drawn from its intention-to-treat analysis (adjusted hazard ratio 0.81, 95% CI 0.63 to 1.04), in which the primary endpoint was not met; its sensitivity, per-protocol and long-term analyses are separate results and are not drawn. ASCOT's primary result is a 3-year landmark percentage (77.1% versus 67.6%, one-sided P value), not a median, so it is listed in the data table but not plotted with the medians. Eras (enrolment between 2006 and 2018), follow-up (for example a median of 46.5 months in PRODIGE 12 and 60 months in the BILCAP primary analysis) and the use of adjusted hazard ratios also differ. Compare only within each trial's own two arms; the lane makes no recommendation and no claim that regimens are equivalent.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

020406080OS, median (months)201820182019BCATPRODIGE 12-ACCORD 18BILCAP
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
OS, median, across 3 trials in this lane. BCAT (2018): experimental 62.3 months vs control 63.8 months; PRODIGE 12-ACCORD 18 (2018): experimental 75.8 months vs control 50.8 months; BILCAP (2019): experimental 51.1 months vs control 36.4 months.
Data table
TrialYearEndpointControl (months)Experimental (months)
BCAT2018OS63.8 months62.3 months
PRODIGE 12-ACCORD 182018OS50.8 months75.8 months
BILCAP2019OS36.4 months51.1 months
ASCOT2023OS (not plotted: different measure or time point)67.6 %77.1 %

Advanced biliary tract cancer after gemcitabine-based chemotherapy: second line

Adults with locally advanced or metastatic biliary tract cancer (cholangiocarcinoma, gallbladder or ampullary carcinoma) that had progressed on first-line cisplatin and gemcitabine, with ECOG performance status 0 or 1 · Relapsed / refractory · endpoint: OS

Results are not comparable across trials.

This lane holds a single randomised trial, ABC-06, so nothing is compared across trials. Its control arm was active symptom control alone, not another chemotherapy; the trial was open label, with 81 people in each arm. The plotted result is the final overall survival analysis (median follow-up 21.7 months), and its hazard ratio (0.69; 95% CI 0.50 to 0.97) is adjusted for the stratification factors (platinum sensitivity, serum albumin concentration and disease stage). Grade 3 to 5 adverse events were more frequent with FOLFOX (69% versus 52%). ClarIDHy, a placebo-controlled trial in IDH1-mutant cholangiocarcinoma with progression-free survival as its primary endpoint, is a different population and endpoint and is not part of this lane, and the single-arm studies of biomarker-matched drugs in previously treated disease have no control arm. Compare only the trial's own two arms; the lane makes no recommendation about second-line treatment beyond that single question.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

0246810OS, median (months)2021ABC-06
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
OS, median, across 1 trial in this lane. ABC-06 (2021): experimental 6.2 months vs control 5.3 months.
Data table
TrialYearEndpointControl (months)Experimental (months)
ABC-062021OS5.3 months6.2 months

FGFR2-rearranged cholangiocarcinoma, first line: FGFR inhibitor versus gemcitabine and cisplatin

Adults with unresectable, locally advanced or metastatic cholangiocarcinoma with an FGFR2 fusion or other rearrangement and no earlier systemic therapy for advanced disease · Metastatic · endpoint: PFS

Results are not comparable across trials.

Two randomised first-line trials are shown, each against gemcitabine plus cisplatin. Both ended early and are small, so neither settles the question. FIGHT-302 (pemigatinib 13.5 mg daily; 167 randomised, 83 and 84) closed early with a reduced sample size and event count, so its P value is nominal; progression-free survival by independent central review was 8.3 versus 6.8 months (hazard ratio 0.58; 95% CI 0.39 to 0.87). It allowed one cycle of gemcitabine-cisplatin before randomisation and crossover to pemigatinib after progression on chemotherapy, and its overall survival (24.4 versus 25.0 months) is confounded by that crossover and is not drawn. PROOF 301 (infigratinib 125 mg on days 1 to 21 of 28-day cycles; 48 randomised, 29 and 19, in a 2:1 ratio) was terminated early for poor accrual against a target of about 300; its report states that inferential statistical analyses could not be carried out, it gives no hazard ratio, the confidence intervals of its medians by blinded independent central review run to not evaluable, and the medians were 7.4 months with infigratinib and 8.0 months with gemcitabine-cisplatin. It excluded gallbladder and ampullary carcinoma and, after an amendment, allowed up to one cycle of gemcitabine-based chemotherapy before randomisation. The drugs, doses, randomisation ratios and sizes differ, so the two results are not averaged, are not read as a class effect, and the lane does not compare pemigatinib with infigratinib. The single-arm FGFR2 studies in previously treated disease have no control arm and are not in this lane. Compare only within each trial's own two arms; the lane makes no recommendation and no claim of equivalence.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

0246810PFS, median (months)20242024FIGHT-302PROOF 301
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
PFS, median, across 2 trials in this lane. FIGHT-302 (2024): experimental 8.3 months vs control 6.8 months; PROOF 301 (2024): experimental 7.4 months vs control 8 months.
Data table
TrialYearEndpointControl (months)Experimental (months)
FIGHT-3022024PFS6.8 months8.3 months
PROOF 3012024PFS8 months7.4 months

Standard of care, then and now

How the usual sequence of treatment changed for one setting. Choose a point in time; each step links to the milestone behind it.

Advanced biliary tract cancer: systemic regimens with a trial result or a regulatory action

Adults with locally advanced unresectable or metastatic biliary tract cancer. Steps are options by treatment setting or biomarker, not a required order, and not every authorised option is shown. · Metastatic

  1. Immunotherapy11st line: gem-cis + durvalumabTOPAZ-1: durvalumab added to gemcitabine and cisplatin (2022)
  2. Immunotherapy21st line: gem-cis + pembrolizumabKEYNOTE-966: pembrolizumab added to gemcitabine and cisplatin (2023)
  3. Chemotherapy3After gemcitabine-cisplatin: FOLFOXABC-06: second-line FOLFOX added to active symptom control (2021)
  4. Targeted4FGFR2 fusion, after therapy: FGFR inhibitorFGFR2 inhibitors: pemigatinib (FIGHT-202) and futibatinib (2020)
  5. Targeted5IDH1 mutation, after therapy: ivosidenibClarIDHy: ivosidenib versus placebo in IDH1-mutant cholangiocarcinoma (2020)
  6. Immunotherapy6HER2 IHC 3+, after therapy: zanidatamabHER2-positive disease: zanidatamab and trastuzumab deruxtecan (2023)
  7. Targeted7BRAF V600E: dabrafenib + trametinibBRAF V600E, NRG1 fusions and tissue-agnostic options (2020-2026) (2020)
  8. Immunotherapy8NRG1 fusion, after therapy: zenocutuzumabBRAF V600E, NRG1 fusions and tissue-agnostic options (2020-2026) (2020)
2026. Based only on the cited milestones. Gem-cis: gemcitabine and cisplatin. Zanidatamab (HER2 IHC 3+): FDA accelerated 20 Nov 2024, EU conditional 27 Jun 2025. Zenocutuzumab (NRG1): FDA 8 May 2026. Lirafugratinib (FGFR2): FDA 23 Sep 2026.
Text version
  • 2010: 1st line: gemcitabine + cisplatin
  • 2021: 1st line: gemcitabine + cisplatin → After gemcitabine-cisplatin: FOLFOX → FGFR2 fusion, after therapy: pemigatinib → IDH1 mutation, after therapy: ivosidenib
  • 2023: 1st line: gem-cis + durvalumab → 1st line: gem-cis + pembrolizumab → After gemcitabine-cisplatin: FOLFOX → FGFR2 fusion, after therapy: FGFR inhibitor → IDH1 mutation, after therapy: ivosidenib → BRAF V600E: dabrafenib + trametinib
  • 2026: 1st line: gem-cis + durvalumab → 1st line: gem-cis + pembrolizumab → After gemcitabine-cisplatin: FOLFOX → FGFR2 fusion, after therapy: FGFR inhibitor → IDH1 mutation, after therapy: ivosidenib → HER2 IHC 3+, after therapy: zanidatamab → BRAF V600E: dabrafenib + trametinib → NRG1 fusion, after therapy: zenocutuzumab

Reversals and dead ends

Approaches that were abandoned or withdrawn, and trials whose results were later questioned, are part of the story.

  • 2018

    Photodynamic therapy: small positive trials, then PHOTOSTENT-02

    Two small randomised trials in the 2000s reported longer survival when photodynamic therapy, a light-activated treatment inside the bile duct, was added to a stent. A larger randomised trial of 92 people later reported shorter survival with it: about six months against about ten. Its authors concluded it cannot be recommended outside clinical trials.

    PHOTOSTENT-02 (Pereira 2018): open-label randomised phase 3, 92 patients with locally advanced or metastatic biliary tract cancer and adequate drainage, porfimer sodium photodynamic therapy plus stenting vs stenting alone (46 per group); recruitment stopped on data-monitoring advice. Median OS 6.2 vs 9.8 months (HR 1.56, 95% CI 1.00-2.43, p=0.048); PFS 3.4 vs 4.3 months (HR 1.43, 0.93-2.18). Earlier randomised trials (Ortner 2003; Zoepf 2005, 32 patients) reported longer survival.

  • 2018

    Endoscopic versus percutaneous drainage before perihilar resection

    A Dutch randomised trial compared two ways of draining a blocked bile duct before liver surgery for perihilar bile duct cancer. It closed early because more people died in the percutaneous drainage group: eleven of 27 against three of 27. Severe drainage complications were similar. The trial was small, and the authors urged caution.

    Coelen 2018: multicentre randomised trial at four Dutch academic centres; 54 patients with potentially resectable perihilar cholangiocarcinoma randomised 1:1 to endoscopic (27) or percutaneous transhepatic (27) biliary drainage. Closed prematurely for higher mortality with percutaneous drainage: 11 (41%) vs 3 (11%) with endoscopic, RR 3.67 (95% CI 1.15-11.69), p=0.03. Primary outcome, severe preoperative drainage-related complications (percutaneous vs endoscopic): 63% vs 67%, RR 0.94 (95% CI 0.64-1.40).

  • 2018

    BCAT, PRODIGE 12: adjuvant gemcitabine trials, no significant benefit

    Two randomised trials tested gemcitabine-based chemotherapy after surgery for bile duct and related cancers: gemcitabine alone in one, gemcitabine with oxaliplatin in the other. Neither found a significant survival difference compared with observation or surveillance, and severe side effects were more frequent with the combination.

    BCAT (Ebata 2018): 225 patients with resected bile duct cancer, gemcitabine (117) vs observation (108): median OS 62.3 vs 63.8 months, HR 1.01 (95% CI 0.70-1.45), p=0.964. PRODIGE 12 (Edeline 2019): 196 patients, 12 cycles of gemcitabine plus oxaliplatin vs surveillance: median RFS 30.4 vs 18.5 months, HR 0.88 (0.62-1.25), p=0.48; OS 75.8 vs 50.8 months, HR 1.08 (0.70-1.66).

  • 2024

    FGFR2 first-line trials end early; infigratinib approval withdrawn

    Two randomised trials compared an FGFR2 inhibitor with gemcitabine and cisplatin as first treatment. Both closed early, so neither settled the question: FIGHT-302 reported longer progression-free survival with pemigatinib, while PROOF 301 enrolled only 48 people. In 2024 the FDA withdrew infigratinib's accelerated approval at the sponsor's request.

    FIGHT-302 (phase 3, open-label, 1:1; pemigatinib 83 vs gemcitabine-cisplatin 84; closed early after a change in standard of care): median PFS 8.3 vs 6.8 months, HR 0.58 (95% CI 0.39-0.87), nominal P=.0078; ORR 47% vs 15%; median OS 24.4 vs 25.0 months, with crossover. PROOF 301 (2:1; infigratinib 29 vs 19; terminated for poor accrual): median PFS by BICR 7.4 vs 8.0 months; no inferential analysis.

Today and tomorrow

Standard of care today

Adjuvant

Five randomised trials of adjuvant therapy after resection were read for this page (abstracts only; full texts not opened), with differing results. BILCAP (UK; 447 people; capecitabine versus observation) did not meet its primary endpoint in the intention-to-treat analysis:

  • median overall survival was 51.1 months with capecitabine and 36.4 months with observation (adjusted hazard ratio 0.81
  • 95% CI 0.63 to 1.04
  • p=0.097). The abstract reports a sensitivity analysis (hazard ratio 0.71
  • 95% CI 0.55 to 0.92) and a prespecified per-protocol analysis (hazard ratio 0.75
  • 95% CI 0.58 to 0.97) that favoured capecitabine, and its authors wrote that capecitabine could be considered a standard of care. ASCOT (JCOG1202
  • Japan
  • 440 people
  • S-1 versus observation) reported 3-year overall survival of 77.1% with S-1 and 67.6% with observation (adjusted hazard ratio 0.69
  • 95% CI 0.51 to 0.94
  • one-sided p=0.0080). Two trials of gemcitabine-based regimens found no significant difference: BCAT (225 people with resected bile duct cancer
  • gemcitabine versus observation
  • overall survival hazard ratio 1.01
  • 95% CI 0.70 to 1.45) and PRODIGE 12 (196 people
  • GEMOX versus surveillance
  • relapse-free survival hazard ratio 0.88
  • 95% CI 0.62 to 1.25). ACTICCA-1 (adjuvant gemcitabine and cisplatin
  • the 2015 design paper's title names observation as the comparator, while the results abstract reports capecitabine) published its results online on 24 October 2026. According to the abstract, 591 people were randomised and median follow-up was 4.8 years
  • median disease-free survival was 25.7 months with gemcitabine and cisplatin and 17.9 months with capecitabine (log-rank p=0.097). Because the proportional-hazards assumption was violated, the prespecified analysis used restricted mean survival time: on average 2.6 additional disease-free months over the first 3 years (95% CI 0.6 to 4.68
  • p=0.011), with no significant improvement in overall survival (restricted mean survival time p=0.076
  • medians 58.1 versus 48.2 months). The abstract concludes that gemcitabine and cisplatin did not statistically improve disease-free survival compared with capecitabine and should not be applied as standard of care in the adjuvant setting. The abstract of the 2019 ASCO guideline recommends offering adjuvant capecitabine for 6 months after resection of biliary tract cancer and says people with extrahepatic cholangiocarcinoma or gallbladder cancer and a microscopically positive margin may be offered chemoradiation
  • the full guideline text was not opened. The ESMO guideline (2023) and the Pan-Asian adapted ESMO guideline (2024) are cited by identifier only. ARTEMIDE-Biliary01 is listed under trials to watch
Primrose JN, Fox RP, Palmer DH, Malik HZ, Prasad R, Mirza D, et al 2019Nakachi K, Ikeda M, Konishi M, Nomura S, Katayama H, Kataoka T, et al 2023Ebata T, Hirano S, Konishi M, Uesaka K, Tsuchiya Y, Ohtsuka M, et al 2018Edeline J, Benabdelghani M, Bertaut A, Watelet J, Hammel P, Joly JP, et al 2019Guideline: Shroff RT, Kennedy EB, Bachini M, Bekaii-Saab T, Crane C, Edeline J, et al 2019Guideline: Vogel A, Bridgewater J, Edeline J, Kelley RK, Klümpen HJ, Malka D, et al 2023Guideline: Chen LT, Vogel A, Hsu C, Chen MH, Fang W, Pangarsa EA, et al 2024Stein A, Arnold D, Bridgewater J, Goldstein D, Jensen LH, Klümpen HJ, et al 2015Stein A, Bridgewater J, Klümpen HJ, Gruenberger T, Shannon J, Fornaro L, et al 2026ClinicalTrials 2026
Neoadjuvant

One randomised comparison of neoadjuvant therapy was read:

  • ZSAB-neoGOLP (New England Journal of Medicine 2026
  • phase 2-3
  • 178 people with resectable high-risk intrahepatic cholangiocarcinoma) compared neoadjuvant GOLP (gemcitabine-oxaliplatin with toripalimab and lenvatinib) followed by resection with resection alone
  • all people received adjuvant capecitabine. At an interim analysis (median follow-up 16.9 months) median event-free survival was 18.0 months versus 8.7 months (P<0.001), and overall survival at 24 months was 79% versus 61% (hazard ratio for death 0.43
  • 95% CI 0.23 to 0.79
  • P=0.005, which did not meet the trial's significance criterion of two-sided alpha 0.0019). The abstract states that no neoadjuvant treatment had been considered standard therapy for this group
  • only the abstract was read. PURITY, a randomised phase 2/3 trial of preoperative chemotherapy in resectable biliary tract cancer, is listed under trials to watch
Shi GM, Huang XY, Liang F, Liang X, Dong R, Ye QH, et al 2026ClinicalTrials 2026
Localised
Two non-randomised studies of liver transplantation after neoadjuvant chemoradiation for selected people with perihilar cholangiocarcinoma were read. A review from one centre (Rea 2005) reported that 71 people entered the transplant protocol and 38 underwent transplantation, with 5-year patient survival of 82% after transplantation and 21% after resection (P=0.022); the groups were not randomised. An analysis of data collected from 12 US centres (Darwish Murad 2012; 287 people treated with neoadjuvant therapy intended to be followed by transplantation) reported intention-to-treat survival of 68% at 2 years and 53% at 5 years, with 71 people dropping out before transplantation. Abstracts only; no guideline text was opened.
Rea DJ, Heimbach JK, Rosen CB, Haddock MG, Alberts SR, Kremers WK, et al 2005Darwish Murad S, Kim WR, Harnois DM, Douglas DD, Burton J, Kulik LM, et al 2012
Metastatic
Cisplatin plus gemcitabine is the chemotherapy backbone in the first-line trials read. ABC-02 (410 people) reported median overall survival of 11.7 versus 8.1 months against gemcitabine alone (hazard ratio 0.64; 95% CI 0.52 to 0.80). Adding durvalumab in TOPAZ-1 (685 people with previously untreated unresectable or metastatic disease, or recurrent disease) gave an overall survival hazard ratio of 0.80 (95% CI 0.66 to 0.97; P=0.021), with estimated 24-month survival of 24.9% versus 10.4%. Adding pembrolizumab in KEYNOTE-966 (1069 people) gave median overall survival of 12.7 versus 10.9 months (hazard ratio 0.83; 95% CI 0.72 to 0.95; one-sided p=0.0034). In SWOG S1815 (441 analysable people), adding nab-paclitaxel did not improve overall survival (median 14.0 versus 13.6 months; hazard ratio 0.91; 95% CI 0.72 to 1.14; P=.41). All figures are from the abstracts (two published corrections to the KEYNOTE-966 paper were not read); no guideline text was opened.
Valle J, Wasan H, Palmer DH, Cunningham D, Anthoney A, Maraveyas A, et al 2010Oh DY, Ruth He A, Qin S, Chen LT, Okusaka T, Vogel A, et al 2022Kelley RK, Ueno M, Yoo C, Finn RS, Furuse J, Ren Z, et al 2023Shroff RT, King G, Colby S, Scott AJ, Borad MJ, Goff L, et al 2025
Metastatic

Regulatory actions read for first-line immunotherapy. FDA:

  • the letter for BLA 761069/S-035 (signed 2 September 2022) provides for Imfinzi (durvalumab) with gemcitabine and cisplatin in adults with locally advanced or metastatic biliary tract cancer, and the letter for BLA 125514/S-144 (signed 31 October 2023) adds Keytruda (pembrolizumab) with gemcitabine and cisplatin for patients with locally advanced unresectable or metastatic biliary tract cancer, citing the final analysis of KEYNOTE-966
  • neither letter uses the word accelerated. EMA procedural-steps tables: Imfinzi variation II/0046 (extension to first-line treatment of adults with unresectable or metastatic biliary tract cancer) shows an opinion dated 10/11/2022 and a Commission decision dated 16/12/2022
  • Keytruda variation II/0138 (first-line treatment of locally advanced unresectable or metastatic biliary tract carcinoma in adults) shows an opinion dated 09/11/2023 and a Commission decision dated 11/12/2023
U 2022U 2023European Medicines Agency 2026European Medicines Agency 2026
Relapsed / refractory

For second-line treatment without a matched alteration, ABC-06 (162 people, UK; progression on cisplatin and gemcitabine; the abstract lists no molecular selection) found median overall survival of 6.2 months with active symptom control plus FOLFOX and 5.3 months with active symptom control alone (adjusted hazard ratio 0.69; 95% CI 0.50 to 0.97; p=0.031). Matched options reported in the abstracts read:

  • pemigatinib in FGFR2 fusions or rearrangements (FIGHT-202, a multicohort single-arm phase 2 study
  • objective response in 38 of 107 people, 35.5%
  • 95% CI 26.5 to 45.4)
  • ivosidenib in IDH1-mutant disease (ClarIDHy, randomised and placebo-controlled
  • median progression-free survival 2.7 versus 1.4 months
  • hazard ratio 0.37
  • 95% CI 0.25 to 0.54)
  • zanidatamab in HER2-amplified disease (HERIZON-BTC-01, single-arm
  • confirmed objective response in 33 of 80 people in cohort 1, 41.3%
  • 95% CI 30.4 to 52.8)
  • and dabrafenib with trametinib in BRAF V600E-mutated disease (ROAR, single-arm basket trial, interim analysis
  • investigator-assessed response in 22 of 43 people, 51%
  • 95% CI 36 to 67)
Lamarca A, Palmer DH, Wasan HS, Ross PJ, Ma YT, Arora A, et al 2021Abou-Alfa GK, Sahai V, Hollebecque A, Vaccaro G, Melisi D, Al-Rajabi R, et al 2020Abou-Alfa GK, Macarulla T, Javle MM, Kelley RK, Lubner SJ, Adeva J, et al 2020Harding JJ, Fan J, Oh DY, Choi HJ, Kim JW, Chang HM, et al 2023Subbiah V, Lassen U, Élez E, Italiano A, Curigliano G, Javle M, et al 2020
Relapsed / refractory

FDA actions read for alteration-matched treatment. Letters:

  • pemigatinib (17 April 2020
  • previously treated cholangiocarcinoma with an FGFR2 fusion or other rearrangement
  • accelerated approval)
  • ivosidenib (signed 25 August 2021
  • previously treated cholangiocarcinoma with an IDH1 mutation
  • the letter does not use the word accelerated)
  • futibatinib (effective action date 30 September 2022, restated in a corrected letter signed 5 October 2022
  • accelerated approval)
  • zanidatamab (effective action date 20 November 2024, restated in a corrected letter signed 10 December 2024
  • previously treated, unresectable or metastatic HER2-positive (IHC3+) biliary tract cancer
  • accelerated approval)
  • zenocutuzumab (signed 8 May 2026
  • cholangiocarcinoma with an NRG1 gene fusion and progression on or after prior systemic therapy
  • the letter does not say whether the indication is accelerated)
  • lirafugratinib (NDA 220425, signed 23 September 2026
  • previously treated, unresectable, locally advanced or metastatic cholangiocarcinoma with an FGFR2 fusion or other rearrangement
  • the letter does not use the word accelerated). Notices: the FDA announcement page for lirafugratinib (23 September 2026) does not state the approval type, says the prescribing information is to be posted on Drugs@FDA, and reports objective response of 46% (95% CI 36 to 55) with median duration of response of 11.8 months in 116 people in the single-arm REFOCUS trial
  • the FDA announced on 16 May 2024 the final withdrawal of the accelerated approval of infigratinib (granted 28 May 2021) at the sponsor's request, citing difficulty recruiting for the required first-line confirmatory trial. Whether any accelerated indication has since been converted was not checked
US Food and Drug Administration 2020US Food and Drug Administration 2021US Food and Drug Administration 2022US Food and Drug Administration 2024US Food and Drug Administration 2026US Food and Drug Administration 2026US Food and Drug Administration 2024US Food and Drug Administration 2026
Relapsed / refractory
EMA pages read. Pemazyre (pemigatinib; authorised 26/03/2021), Lytgobi (futibatinib; 04/07/2023) and Ziihera (zanidatamab; 27/06/2025) each state a conditional marketing authorisation, for adults with locally advanced or metastatic cholangiocarcinoma with an FGFR2 fusion or rearrangement after at least one prior line of systemic therapy (pemigatinib and futibatinib) or unresectable locally advanced or metastatic HER2-positive (IHC3+) biliary tract cancer after at least one prior line (zanidatamab). Tibsovo (ivosidenib) has a marketing authorisation issued 04/05/2023; its page lists a monotherapy indication for adults with locally advanced or metastatic cholangiocarcinoma with an IDH1 R132 mutation who were previously treated by at least one prior line of systemic therapy (alongside an acute myeloid leukaemia indication) and states no special authorisation type. Two FDA tissue-agnostic accelerated approvals cover solid tumours: dabrafenib with trametinib for BRAF V600E solid tumours (letter signed 22 June 2022) and trastuzumab deruxtecan for HER2-positive (IHC 3+) solid tumours (5 April 2024). The trastuzumab deruxtecan letter does not name biliary cancer; the dabrafenib letter names biliary tract cancer only in a postmarketing requirement in which the prospectively conducted trial it describes will exclude patients with biliary tract cancer, among other tumour types.
European Medicines Agency 2026European Medicines Agency 2025European Medicines Agency 2026European Medicines Agency 2023US Food and Drug Administration, Center for Drug Evaluation and Research 2022US Food and Drug Administration 2024

Trials to watch

  • ARTEMIDE-Biliary01 NCT06109779

    After macroscopically complete resection of biliary tract cancer (intrahepatic or extrahepatic cholangiocarcinoma or muscle-invasive gallbladder cancer), does rilvegostomig added to investigator's choice of chemotherapy differ from placebo with chemotherapy in recurrence-free survival (the registry's primary outcome)? Registry record: phase 3, randomised, double-blind, placebo-controlled; 760 enrolled (actual); sponsor AstraZeneca. Registry status: active, not recruiting (status verified 2026-10); no results posted.

  • JZP598-302 NCT06282575

    In advanced HER2-positive biliary tract cancer, does zanidatamab with standard-of-care therapy differ from standard-of-care therapy alone in progression-free survival in participants with HER2 immunohistochemistry 3+ tumours (the registry's primary outcome)? Registry record: phase 3, randomised, open-label (per the official title); 286 estimated; sponsor Jazz Pharmaceuticals; the registry summary describes first-line zanidatamab with cisplatin and gemcitabine, with or without a PD-1 or PD-L1 inhibitor (durvalumab or pembrolizumab, where approved). The registry organisation study ID is JZP598-302, and the official title is the same as that of the randomised confirmatory study JZP598-302 named in the zanidatamab FDA approval letter. Registry status: recruiting (status verified 2026-09); no results posted.

  • DESTINY-BTC01 NCT06467357

    In previously untreated, locally advanced or metastatic HER2-expressing biliary tract cancer, does trastuzumab deruxtecan with rilvegostomig, or trastuzumab deruxtecan alone, differ from gemcitabine, cisplatin and durvalumab in overall survival in the HER2 immunohistochemistry 3+ full analysis set (the registry's randomised-portion primary outcome; the safety run-in has its own primary outcome)? Registry record: phase 3, randomised; 620 estimated; sponsor AstraZeneca. Registry status: recruiting (status verified 2026-10); no results posted.

  • FIRST-308 NCT05948475

    In cholangiocarcinoma with a documented FGFR2 fusion or rearrangement after at least one line of chemotherapy and exactly one FDA-approved FGFR inhibitor, does tinengotinib differ from physician's choice of treatment in progression-free survival by blinded independent central review (the registry's Part B primary outcome; the Part A primary outcomes concern adverse events)? Registry record: phase 3, randomised, global multicentre; 200 estimated; sponsor TransThera Sciences. Registry status: active, not recruiting (status verified 2026-07; primary completion listed as 2026-12, estimated); no results posted.

  • PURITY NCT06037980

    In resectable, non-metastatic biliary tract cancer at high risk of recurrence, does preoperative cisplatin, gemcitabine and nab-paclitaxel differ from upfront surgery in the 12-month progression-free survival rate (phase II part) and progression-free survival (phase III part) (the registry's primary outcomes)? Registry record: phase 2/3 (adaptive), randomised; 300 estimated; sponsor Gruppo Oncologico del Nord-Ovest. Registry status: recruiting (status verified 2026-09); no results posted.

Pioneers

References

Show all 122 references

Loading references… Open the list as a page

Every figure was checked against its primary source to the extent possible. Items that could not be fully confirmed are marked “needs review”. Educational history, not medical advice.

Report an error

Links open the primary source. A flag marked ⚑ needs review means a value or claim has not yet been independently confirmed.

2023TreatmentImmunotherapy

majorRelapsed / refractoryMetastatic

HER2-positive disease: zanidatamab and trastuzumab deruxtecan

Some biliary tract cancers carry extra copies of the HER2 gene or high HER2 protein levels. In single-arm studies of previously treated patients, HER2-directed drugs (zanidatamab, trastuzumab deruxtecan) produced responses in roughly one in four to one in two people. The FDA gave zanidatamab accelerated approval in 2024; a randomised confirmation trial is required.

Single-arm studies in previously treated HER2-positive biliary tract cancer. HERIZON-BTC-01, zanidatamab: cohort 1 (IHC 2+/3+, 80 patients) confirmed ORR 41.3% (95% CI 30.4-52.8); IHC 3+ (label, 62 patients) ORR 52% (39-65), median DOR 14.9 months. DESTINY-PanTumor02 biliary cohort, trastuzumab deruxtecan: ORR 22.0% (10.6-37.6); central IHC 3+ 56.3% (16 patients). HERB: ORR 36.4% (22 patients).

Standard before

The MyPathway report states that systemic therapies for metastatic biliary tract cancer were few and that patients had a median overall survival of less than 1 year.

What changed

The FDA granted accelerated approval to zanidatamab-hrii on 20 November 2024 for previously treated, unresectable or metastatic HER2-positive (IHC3+) biliary tract cancer as detected by an FDA-approved test; the EMA granted a conditional marketing authorisation on 27 June 2025 for the same IHC3+ population. Separately, the FDA granted accelerated approval to trastuzumab deruxtecan for HER2-positive (IHC 3+) solid tumours on 5 April 2024.

Why it mattered

These studies defined HER2-positive biliary tract cancer as a treatable molecular subset after gemcitabine-based therapy and provided the response-rate data behind an accelerated approval for zanidatamab; the DESTINY-PanTumor02 biliary cohort adds a second HER2-directed drug with a different response profile.

Caveats

  • All four studies are single-arm. Response rates are not comparisons with a control or with each other: HER2 status was defined differently (zanidatamab: amplification by in-situ hybridisation, with cohort 1 IHC 2+ or 3+; trastuzumab deruxtecan in DESTINY-PanTumor02: IHC 3+ or 2+ by local or central testing; HERB: central IHC 3+ or IHC 2+ with positive ISH), and the FDA and EMA zanidatamab indications are limited to IHC 3+.
  • The FDA zanidatamab approval is accelerated. The approval letter requires completion of the randomised trial Study JZP598-302 of zanidatamab with standard of care versus standard of care alone in HER2-positive (IHC 3+) disease, with a timetable listing completion in March 2029; the label version effective 24 August 2026 still states accelerated approval. The EMA authorisation is conditional.
  • The IHC 3+ response rate and duration of response for zanidatamab are printed in the FDA label, whose clinical studies section gives no data cutoff, so they may reflect a later cutoff than the 2023 paper.
  • The DESTINY-PanTumor02 biliary figures are for one of seven cohorts, and the 22.0% rate is investigator-assessed (26.8% by independent central review).
  • Trastuzumab deruxtecan carries a risk of interstitial lung disease: DESTINY-PanTumor02 reported adjudicated drug-related ILD in 10.5% of 267 patients, including three deaths across the study, and HERB reported ILD in 8 patients (25.0%), two of them grade 5.
  • The trastuzumab deruxtecan tissue-agnostic approval is defined by the HER2 IHC 3+ biomarker, not by tumour type, and is accelerated.