Lirafugratinib Is Approved as Lyrfigtu, the First FGFR2-Only Drug for Bile Duct Cancer

For most people with bile duct cancer, the test result that changes everything arrives weeks after the diagnosis. The tumor sample goes off for genomic profiling, and in somewhere between 1 in 10 and 1 in 5 cases of the kind that starts inside the liver, the report comes back with 2 words that had no approved treatment before 2020: FGFR2 fusion. It means a piece of the FGFR2 gene has been welded onto an unrelated gene, and the hybrid protein it makes is telling the cancer to grow without stopping. It also means there are pills designed to switch that signal off.

3 of those pills have been approved in the United States since 2020, and 1 was pulled. All 3 block FGFR2 along with its close relatives, which is where many of their side effects come from. On September 23, 2026, 2 days before its deadline, the FDA approved a fourth: lirafugratinib, sold as Lyrfigtu, the first one built to hit FGFR2 and nothing else.

FGFR2 Fusions in Cholangiocarcinoma, and Why the Tumor Test Decides Everything

Cholangiocarcinoma forms in the bile ducts, the thin tubes that carry bile out of the liver. About 8,000 people in the United States are diagnosed with it each year, a figure the American Cancer Society notes is probably an undercount because the cancer is hard to diagnose and sometimes gets misclassified (American Cancer Society, 2026). Most cases are found late. Surgery is the only route to a cure, and most patients are not candidates by the time they are diagnosed.

Advanced disease is treated first with gemcitabine and cisplatin chemotherapy, usually now with an immunotherapy added. What happens after that depends on the tumor's genes. The FDA's own review of the first FGFR drug put the share of intrahepatic cholangiocarcinomas carrying an FGFR2 fusion at 10 to 20% (FDA, 2020). Other subgroups have their own drugs: IDH1 mutations (ivosidenib) and strong HER2 overexpression (zanidatamab) among them. A patient whose tumor was never profiled has no way to know which door is open.

The FDA approved pemigatinib (Pemazyre) on April 17, 2020, the first targeted therapy for cholangiocarcinoma. Infigratinib (Truseltiq) followed on May 28, 2021, and futibatinib (Lytgobi) on September 30, 2022. All 3 were accelerated approvals, granted on how many tumors shrank in a single-arm trial rather than on proof that patients lived longer. That kind of approval comes with a string attached, written into both of the labels still in use in the same words.

“Continued approval for this indication may be contingent upon verification and description of clinical benefit in a confirmatory trial(s).”

Pemazyre and Lytgobi prescribing information, U.S. Food and Drug Administration

Truseltiq could not meet it. Its confirmatory trial was set up in newly diagnosed patients, and the sponsor could not enroll enough of them. The FDA finalized the withdrawal on May 16, 2024, recording the sponsor's reasoning that continuing to sell the drug for its approved second-line use was no longer commercially reasonable (FDA, 2024). The drug left the market over enrollment and economics, with no finding that it had stopped working or become unsafe.

Lirafugratinib nearly had a quieter version of the same problem. Its developer, Relay Therapeutics, had finished enrolling the pivotal cholangiocarcinoma cohort by October 2023, and that same month it paused its near-term commercial preparations for the disease, saying it would pursue a broader use across tumor types and put its own focus on other drugs (Relay Therapeutics, 2023). The drug then sat with a finished pivotal cohort and no company preparing to launch it in bile duct cancer until December 3, 2024, when Relay licensed worldwide rights to Elevar Therapeutics, a New Jersey subsidiary of Korea's HLB Group, in a deal worth up to $500 million (Relay Therapeutics, 2024).

Why Lirafugratinib Was Built to Block FGFR2 Alone

FGFR2 is one of 4 closely related receptors, and the drugs before lirafugratinib block several of them at once. That matters because the relatives do jobs of their own. FGFR1 helps the body regulate phosphate, so blocking it drives blood phosphate up; FGFR4 affects the gut, so blocking it causes diarrhea. On the older drugs these are not rare events. By their FDA labels, 93% of patients on pemigatinib and 88% on futibatinib had phosphate above the normal range (FDA prescribing information). The scientists who designed lirafugratinib named the consequence plainly.

“The toxicity of these drugs frequently leads to dose reduction or interruption of treatment.”

Schönherr et al., Proceedings of the National Academy of Sciences, 2024

The pockets where these drugs bind look nearly identical across the 4 receptors in still images, which is why earlier attempts at an FGFR2-only drug struggled. Relay and D. E. Shaw Research ran long computer simulations of how the receptors actually move and found a flexible loop that shifts differently in FGFR2 than in FGFR1. Lirafugratinib was built to latch onto that loop and bond to it permanently. In lab tests it is about 250 times more selective for FGFR2 than FGFR1 and about 5,000 times more selective than FGFR4, and its designers reported that in early patients the effective dose did not cause clinically significant high phosphate or diarrhea (Schönherr et al., 2024). In the larger safety data behind the approval, high phosphate was still reported in 21% of patients, though almost always mild, and diarrhea was among the most common side effects (FDA label, 2026).

ReFocus Trial Results for Lirafugratinib in FGFR2-Fusion Cholangiocarcinoma

The approval rests on 1 cohort of the ReFocus trial (NCT04526106): 116 adults with advanced FGFR2 fusion or rearrangement cholangiocarcinoma who had already had at least 1 treatment but had never taken an FGFR drug. They took 70 mg once a day. Their median age was 57, and 61% were women. An independent committee, not the treating doctors, judged the scans, and 114 patients were in that analysis (Hollebecque et al., 2026). The FDA's approval notice reports the result as a 46% response rate (95% CI 36 to 55) with responses lasting a median 11.8 months (FDA, 2026).

Set next to the 3 drugs that came before it, the numbers line up in a way that is hard to miss, and just as easy to overread. Each bar below comes from a different single-arm trial with its own patients, sites and years, and none of these drugs has been tested against another. The chart shows each drug's own result, not a race.

Lirafugratinib Side Effects Move to the Hands, Feet and Mouth

Sparing FGFR1 and FGFR4 does not make this a gentle drug, and the trial data say so directly. Blocking FGFR2 itself affects the skin, nails and lining of the mouth, and in ReFocus those were the dominant problems. Severe hand-foot syndrome, the painful redness and peeling of the palms and soles doctors call palmar-plantar erythrodysesthesia, hit 32.8% of patients. Severe mouth sores hit 12.1%. 3 out of 4 patients needed their dose lowered and more than 4 in 5 needed a pause at some point (Hollebecque et al., 2026).

The number that tempers those is the one at the end: only 4.3% stopped the drug because of side effects, and quality of life, measured by a standard questionnaire, held steady. The pattern is a drug that most people can stay on, with frequent adjustments. Anyone starting it should expect the care team to watch their hands, feet and mouth closely from the first weeks, and should report tenderness early rather than waiting for peeling.

The FDA label adds 3 warnings that the conference data did not put front and center: eye toxicity, high blood phosphate with soft tissue mineralization (calcium and phosphate settling into tissues such as blood vessels and skin), and harm to a developing baby (FDA, 2026). Eye problems are a known effect of the whole FGFR class, and the Pemazyre and Lytgobi labels call for eye exams before starting and regularly during treatment. Lyrfigtu's full prescribing information, which the FDA says will be posted on Drugs@FDA, sets its own monitoring schedule, and anyone who could become pregnant, or whose partner could, should ask about contraception before the first dose.

When Lirafugratinib Stops Working, Futibatinib May Still Help

Every FGFR drug in this disease eventually meets resistance, usually because the cancer develops a new mutation in FGFR2 that stops the drug from binding. The concern with each new drug has been whether it simply trades one set of escape mutations for another, leaving nothing to try next. 2 papers published this year give an unexpectedly hopeful answer.

At Gustave Roussy in France, researchers tracked 30 patients treated with lirafugratinib. In 11 of 16 patients whose cancer became resistant, the escape route ran through 2 specific spots in FGFR2, positions M538 and L618, while the gatekeeper and molecular-brake mutations that typically defeat futibatinib were rare. The 2 drugs appear to leave different doors open. 3 patients switched to futibatinib after lirafugratinib stopped working and had prolonged responses, and the authors concluded the 2 can be used in sequence when the specific resistance mutation is known (Facchinetti et al., 2026). A Massachusetts General Hospital team found a similar split in 28 patients and also reported that lirafugratinib worked in some people whose cancer had already outlasted an older FGFR drug (Ellis et al., 2026).

What the Lyrfigtu Approval Covers, and What It Leaves Open

The label covers adults whose cholangiocarcinoma cannot be removed with surgery or has spread, who have already had treatment, and whose tumor carries an FGFR2 fusion or other rearrangement. The dose is 70 mg by mouth once a day until the cancer grows or side effects become too much. The application had priority review plus breakthrough therapy and orphan drug designations, and it went through the FDA's Real-Time Oncology Review program, which lets a company send finished data in pieces before the full application is filed (FDA, 2026).

The FDA's notice calls this an approval, with no mention of accelerated approval. Its notices for pemigatinib, infigratinib and futibatinib each announced an accelerated approval in the headline, and Relay had described the pivotal cohort in 2024 as designed to support one (Relay Therapeutics, 2024). The full prescribing information is where any confirmatory-trial condition would appear, and the lirafugratinib drug page will carry it once the FDA posts the label.

Approval does not change first-line treatment. It adds a third FGFR option after chemotherapy for patients with a fusion, one with a different side-effect profile from the 2 already there, and, if the resistance research holds up, one that may fit before or after futibatinib rather than competing with it. Elevar's European application, filed September 15, 2026, is still under review. Trials for cholangiocarcinoma continue at sites around the country; the cholangiocarcinoma trials near you page ranks them by distance.