Tovorafenib

Pediatric low-grade glioma (pLGG) comprises 30% to 40% of all pediatric CNS tumors and includes WHO grade 1 and 2 glioma, glioneuronal, and neuronal tumors, which the 2021 WHO classification places under pediatric-type diffuse low-grade glioma, circumscribed astrocytic tumors, and glioneuronal and neuronal tumors. The disease is driven predominantly by RAS/MAPK pathway alterations. In a molecularly profiled cohort, 84% of tumors harbored a driver alteration; KIAA1549-BRAF fusion was the most frequent (35%), followed by BRAF p.V600E (17%). In a separate cohort, BRAF V600E was detected in 69 of 405 patients (17%). Company estimates differ in scope: Day One Biopharmaceuticals states both that 50% to 60% of pLGGs are driven by RAF alterations (85% to 90% of them KIAA1549-BRAF fusions) and that approximately 70% of pLGGs in the United States are BRAF-altered. An estimated 15% to 20% of children with neurofibromatosis type 1 (NF1) develop an optic pathway glioma; participants with NF1-driven tumors are excluded from LOGGIC/FIREFLY-2.

No identified publication reports the incidence or prevalence of RAF-altered pLGG requiring first-line systemic therapy, and the component figures below come from different populations, age ranges, and periods; they have not been combined into a published estimate. In the United States, CBTRUS reports an incidence of pilocytic astrocytoma of 0.98 per 100,000 at ages 0 to 19 years and projects 730 new cases in 2025; the LOGGIC/FIREFLY-2 design publication cites 1,200 to 1,500 new pLGG cases per year; and Day One Biopharmaceuticals cites an annual pLGG incidence of 1.3 to 2.1 per 100,000, estimates that approximately 1,100 patients younger than 25 years are newly diagnosed with BRAF-altered pLGG each year, and reports a registry-based prevalence of 26,000 such patients as of January 1, 2017. The proportion of children who need therapy beyond surgery or observation was 165 of 639 (25.8%) starting nonsurgical treatment at diagnosis in the UK CNS9702 cohort (113 chemotherapy, 52 radiotherapy); in the German HIT-LGG-1996 cohort, 184 of 1,031 children proceeded to nonsurgical treatment at diagnosis and 35.2% ultimately received it; and in SIOP-LGG 2004, 1,057 of 3,417 registered, previously untreated patients (30.9%) received chemotherapy.

Overall survival is high, but disease control after first-line chemotherapy is limited. In 4,040 children in SEER, 20-year overall survival was 87% and the 20-year cumulative incidence of death due to glioma was 12%; in an Ontario population-based cohort, 20-year overall survival was 90.1%. Among children with an indication for systemic treatment, the LOGGIC/FIREFLY-2 investigators cite a 10-year overall survival of 94% for pLGG but a 10-year PFS of 44% with standard chemotherapy. In COG A9952 (274 randomized participants), 5-year EFS was 39% with carboplatin and vincristine and 52% with TPCV; in SIOP-LGG 2004 (497 randomized participants), 5-year PFS was 46% with vincristine and carboplatin; and 40% to 50% of patients require subsequent chemotherapy lines. Molecular subgroup modifies prognosis: BRAF V600E tumors had a 10-year PFS of 27% after chemotherapy and radiation therapy versus 60.2% for wild-type tumors and a 5-year PFS of 30.4% after first-line chemotherapy; BRAF p.V600E with CDKN2A deletion had a 10-year PFS of 0% and 10-year overall survival of 41%, whereas tumors with gene fusions or germline NF1 had a 10-year PFS of 67% and overall survival of 98%. Radiation therapy was associated with increased cancer-specific death in SEER (HR 3.9), and upfront radiotherapy was associated with increased late death among Ontario patients with unresectable tumors surviving more than 5 years (HR 3.3).

Late morbidity continues to accumulate beyond 5 years after diagnosis. Among 240 five-year survivors in a St. Jude cohort, the 15-year cumulative incidence was 18% for blindness, 22% for hearing loss, 29% for growth hormone deficiency, 33% for thyroid hormone deficiency, 26% for ACTH deficiency, and 53% for overweight or obesity; monocular legal blindness reached 57% at 15 years among patients with progressive hypothalamic/chiasmatic gliomas, and 61 patients (34%) had an IQ below 85, compared with 16% in the normative sample. After treatment for optic pathway glioma, 18.8% of 117 patients had severe visual impairment or blindness in both eyes at a median of 8.3 years. First-line chemotherapy carries its own toxicity: carboplatin hypersensitivity occurred in 41.9% of patients in one series, with recurrence in 24 of 34 (70.5%) rechallenged patients; visual acuity worsened after chemotherapy in 41% of children with NF1-associated and 39% with sporadic optic pathway glioma; and European recommendations advise central venous catheters, especially for infants and toddlers.

Direct evidence on economic and family burden specific to pLGG is limited. In a US claims and electronic health record study of 154 patients aged 18 years or younger followed for 36 months, 18% to 28% visited an emergency room and 9% to 27% had an inpatient stay in each 6-month interval, mean inpatient stay increased from 5.1 to 8.1 days, and 74% and 56% had prescriptions for anti-infectives and antiemetics. No study of family costs specific to pLGG was identified. Among 86 patients who survived at least 2 years after pediatric neuro-oncology surgery at a US cancer center, 29 (34%) experienced financial toxicity; approximately one third of 254 caregivers of US children with cancer reported that a parent quit or changed work, which was associated with a 13.4-point higher financial burden score (0 to 100 scale) at 1 to 5 years after diagnosis; and 28 Canadian families reported median first-year indirect costs of $9,668 and a median of 55 days of lost income.

No efficacy or safety results for tovorafenib as first-line therapy had been reported as of September 2026. The only first-line trial, LOGGIC/FIREFLY-2 (NCT05566795), is a randomized (1:1), open-label, phase 3 trial comparing tovorafenib once weekly with investigator's choice of 4 standard-of-care chemotherapy regimens (COG vincristine/carboplatin, SIOPe-LGG vincristine/carboplatin, vinblastine, or monthly carboplatin) in participants aged 6 months to 25 years with RAF-altered pLGG requiring first-line systemic therapy. Participants with NF1 or with additional activating molecular alterations are excluded, randomization is stratified by tumor location, fusion versus mutation, CDKN2A status, and infant chiasmatic-hypothalamic glioma, and participants who progress on chemotherapy may cross over to tovorafenib. Tovorafenib is given at 420 mg/m2 (maximum 600 mg), which is higher than the approved 380 mg/m2. Enrollment reached 169 randomized participants at 110 sites by 15 January 2025 and was completed in May 2026 with approximately 400 participants at approximately 140 sites. The primary analysis is expected approximately 12 months after the last enrollment, with topline data expected by mid-2027; the FDA postmarketing requirement specifies an interim report in April 2028 and a final report in April 2032. The sources differ on the primary endpoint: the 2024 design publication specifies IRC-assessed ORR per RANO-LGG, whereas the registry record (which counts minor responses as responses) and the May 2026 sponsor announcement describe ORR per RAPNO-LGG, and the EMA reports that the protocol was amended to RAPNO-LGG after 113 of 400 participants had been recruited. The design assumed a 30% ORR with chemotherapy and provides approximately 85% power to detect a 15% ORR improvement and a PFS hazard ratio of 0.67 (assumed median PFS 4.5 versus 3 years). Key secondary endpoints include PFS, EFS, TTNT, overall survival, and patient-reported outcomes.

All available efficacy data concern relapsed or refractory disease. The single-arm phase 2 FIREFLY-1 trial enrolled 137 participants (77 in the registrational arm 1 and 60 in expansion arm 2) with a median of 3 prior systemic therapies; 83 of 137 (61%) had received MAPK pathway-targeted therapy. At the 5 June 2023 data cutoff, IRC-assessed ORR in arm 1 was 67% by RANO-HGG (69 evaluable participants; the protocol primary endpoint), 51.3% by RAPNO-LGG, and 52.6% by RANO-LGG (76 evaluable participants each), with a median DOR of 13.8 months and median PFS of 13.8 months by RAPNO-LGG; the FDA based accelerated approval on RAPNO-LGG ORR and DOR. At the 10 May 2024 cutoff, RAPNO-LGG ORR was 52.6% with a median DOR of 18.0 months, and RANO-HGG ORR was 71.0% by IRC versus 49.3% by investigators. At the 3-year analysis (6 June 2025; median study duration 40.6 months), RAPNO-LGG ORR was 40 of 76 (53%), median DOR was 19.4 months (24.0 months in MAPK inhibitor-naive and 16.6 months in previously treated participants), median PFS was 16.6 months, 24-month PFS was 32%, and median TTNT was 42.6 months; 38 participants (50%) continued treatment beyond RAPNO-defined progression. Of 39 participants who entered the post-treatment observation period, 30 (77%) were treatment free for at least 12 months and 12 (31%) met criteria for rebound tumor growth, including 4 of 6 with BRAF V600E; 8 were retreated, with a median tumor size change of -38%. Among 35 participants with optic pathway glioma and at least 2 visual acuity assessments, visual acuity was preserved in 80% and improved in 31%. PedsQL Cancer Module scores generally remained stable or improved during the first year, with respondents declining from 51 at baseline to 32 at cycle 13. In the phase 1 PNOC014 study (44 participants aged 1 to 25 years with recurrent or progressive MAPK-altered tumors), 25 of 27 centrally reviewed participants (93%) had complete response, partial response, or stable disease, and 420 mg/m2 was selected as the recommended phase 2 dose.

All safety data also derive from relapsed or refractory populations. The US label has no contraindications and no boxed warning, and no REMS is required. In the pooled label safety population, hemorrhage occurred in 37% (intratumoral hemorrhage 9%; 1 grade 5 tumor hemorrhage), rash in 67% (grade 3 in 12%), photosensitivity in 12%, and increased AST and ALT in 74% and 42%; effects on growth occurred in 46% of 133 FIREFLY-1 participants aged 18 years or younger (35% grade 3 or higher). In FIREFLY-1, the most common adverse reactions were rash (77%), hair color changes (76%), fatigue (55%), viral infection (55%), and vomiting (50%); serious adverse reactions occurred in 45%, and adverse reactions led to dose interruption in 57%, dose reduction in 24%, and permanent discontinuation in 7%. At the 3-year analysis (137 participants), grade 3 or higher TEAEs occurred in 113 (82%) and serious TEAEs in 77 (56%); TRAEs led to discontinuation in 18 (13%) and dose reduction in 45 (33%); decreased growth velocity was reported in 61 (45%; grade 3 or higher in 46, 34%) and intratumoral hemorrhage in 21 (15%); and 5 participants (4%) had grade 5 TEAEs assessed as not related to tovorafenib. Median annualized growth velocity increased from 1.40 cm/year during treatment to 7.23 cm/year after treatment; 67 of 74 participants (91%) with on-treatment growth reduction had growth recovery and 53 (72%) had catch-up growth, and no premature growth plate fusions were observed. In PNOC014, 11 of 12 participants (92%) treated for more than 6 months who were not postpubertal had grade 3 growth suppression. No ocular toxicity, adverse impact on cardiac function, or abnormal weight gain was observed in FIREFLY-1. The EU risk management plan lists growth retardation and intratumoral hemorrhage as important identified risks, decreased fertility as an important potential risk, and long-term safety as missing information.

No tovorafenib price specific to first-line use exists, because no first-line indication is approved. In the United States, the WAC as of 1 January 2025 was $35,272.64 per carton of 100 mg tablets and $8,818.16 per single-use 7-day bottle of oral suspension; WAC does not reflect discounts or rebates. The Institute for Clinical and Economic Review reported an annual list price of $330,720 and an estimated annual net price of $297,648, assuming a 375 mg weekly dose. The only price outside the United States identified is the German company dossier, which reported annual therapy costs of €239,561.93 to €479,123.86 per patient based on the oral suspension; IQWiG judged the lower bound plausible and the upper bound overstated because tablets cost less, and the company estimated 6 to 156 patients in the statutory health insurance target population for relapsed or refractory disease. Treatment requires a BRAF alteration; the FDA-approved companion diagnostic FoundationOne CDx had an initial Medicare payment of $3,500.00 (July 2018 to March 2019).

No cost-effectiveness analysis or budget impact model of tovorafenib in any pLGG population has been published, and all health technology assessment activity concerns relapsed or refractory disease. The EU Joint Clinical Assessment found no direct comparative data and no comparative evidence for the full claimed indication, including comparisons with carboplatin plus vincristine or vinblastine; its only included comparison, an unanchored MAIC versus dabrafenib plus trametinib in BRAF V600E-mutated disease, had effective sample sizes of 5.81 to 14.26. In Germany, the G-BA orphan drug benefit assessment was published on 17 August 2026, with a decision expected in early November 2026. In France, HAS granted early access for patients who progressed after at least one systemic treatment and were previously treated with, or ineligible for, dabrafenib plus trametinib. The NICE appraisal (ID6557) was in scoping. A US federal employee health plan policy (October 2025) covers tovorafenib only for relapsed or refractory pLGG and considers other indications investigational.

Economic evidence for first-line targeted therapy exists only for the comparator dabrafenib plus trametinib in BRAF V600E pLGG. NICE (TA977) recommended dabrafenib plus trametinib for children and young people aged 1 year and over who need systemic treatment, applying a 1.2 severity weight to QALYs and an acceptable ICER of around £30,000 per QALY; the LGG ICERs are confidential. A model for England and Wales reported a probabilistic ICER of £26,606 per QALY gained, with a 70.7% probability of cost-effectiveness at £30,000 per QALY. A Canadian microsimulation of first-line dabrafenib plus trametinib versus chemotherapy (2024 Canadian dollars) estimated a gain of 2.21 QALYs at an incremental cost of CAD 554,769, an ICER of CAD 251,027 per QALY with lifetime treatment and CAD 44,740 per QALY with 2 years of treatment, and concluded that a price reduction would be required.

US OJEMDA net product revenue was $57.2 million in 2024 and $155.4 million in 2025 (4,635 prescriptions in 2025), and 2026 guidance was $225 million to $250 million, with the company describing adoption as standard of care in second-line pLGG. The company expected a mid-2027 FIREFLY-2 readout and a potential first-line approval in 2028. Management projections disclosed for the Servier acquisition assumed development of OJEMDA as a front-line therapy and projected total company net revenue, including other products, royalties, and milestones, of $261 million in 2026, rising to $1,131 million in 2035; OJEMDA-only forecasts were not disclosed. The company estimates approximately 1,100 new BRAF-altered pLGG diagnoses per year in US patients younger than 25 years, an addressable pool of 2,000 to 3,000 relapsed, progressive, or refractory patients per year, and a chronic duration of treatment over many years. Because first-line efficacy, treatment duration, uptake relative to chemotherapy and dabrafenib plus trametinib, first-line pricing, and the size of the eligible population are not established, the budget impact of first-line tovorafenib cannot be estimated from public data as of September 2026.

Tovorafenib is an oral, once-weekly type II RAF inhibitor. In the United States, it received accelerated approval on 23 April 2024 for patients 6 months of age and older with relapsed or refractory pLGG harboring a BRAF fusion or rearrangement or BRAF V600 mutation, based on response rate and duration of response, and it remained on the FDA list of ongoing accelerated approvals in September 2026. In the European Union, a conditional marketing authorisation was granted on 20 April 2026 for the same molecular population after progression on one or more prior systemic therapies. In Japan, approval was obtained on 16 September 2026 for low-grade glioma with a BRAF gene mutation or fusion gene; the indication wording does not specify a line of therapy, although the supporting data came from FIREFLY-1. No marketing authorisation was identified in the United Kingdom, Canada, or Australia, and no supplemental application or EU variation for first-line use was identified. FIREFLY-2 is both the FDA confirmatory trial (final report due April 2032) and the EU specific obligation. First-line use in the United States and European Union is therefore outside the approved indications, and payer policy identified in the United States is limited to relapsed or refractory disease.

What is known for first-line use is limited to indirect and mechanistic evidence. Tovorafenib inhibits RAF monomers and dimers and is active in both BRAF fusion and BRAF V600E tumors, whereas type I BRAF inhibitors are not indicated for BRAF fusion tumors, and no targeted therapy is approved as first-line treatment for BRAF fusion-driven pLGG. The EMA considered that the mechanism of action is not expected to be line-dependent, and the company argues that earlier treatment may improve efficacy; in FIREFLY-1, median DOR was 24.0 months in MAPK inhibitor-naive versus 16.6 months in previously treated participants. What is not known is whether tovorafenib improves response, PFS, visual or functional outcomes, or quality of life compared with first-line chemotherapy; its efficacy and safety at the 420 mg/m2 dose used in FIREFLY-2 relative to the approved 380 mg/m2; the optimal treatment duration and the risk of rebound growth after stopping (31% of participants entering observation in FIREFLY-1); and its long-term effects on growth, adult height, fertility, and late toxicity in treatment-naive children, for which the EU risk management plan lists long-term safety as missing information.

The comparative position is defined by benchmarks from different trials rather than by direct comparison. First-line chemotherapy produced a 5-year EFS of 39% with carboplatin and vincristine in COG A9952 and a 5-year PFS of 46% with vincristine and carboplatin in SIOP-LGG 2004. In the randomized first-line trial of dabrafenib plus trametinib in BRAF V600 pLGG (110 participants), ORR was 47% versus 11%, median PFS was 20.1 versus 7.4 months (HR 0.31), and grade 3 or higher adverse events occurred in 47% versus 94% with carboplatin plus vincristine; the FDA approved this combination for first-line use in BRAF V600E pLGG on 16 March 2023. FIREFLY-2 does not include dabrafenib plus trametinib as a comparator. The EU Joint Clinical Assessment compared tovorafenib with dabrafenib plus trametinib only in relapsed or refractory BRAF V600E disease, using an unanchored MAIC against a dabrafenib plus trametinib study that included small proportions of patients with high-grade glioma or without prior systemic therapy: IRC-assessed PFS by RANO-LGG had a hazard ratio of 4.88 (95% CI 2.14 to 11.14), where values below 1 favor tovorafenib; ORR comparisons were inconclusive; and the odds ratio for grade 3 or higher adverse events was 12.65 (95% CI 2.30 to 69.58). The effective sample sizes were 5.81 to 14.26, prognostic factors including histology, tumor location, and prior chemotherapy were not adjusted for, and the assessors stated that the estimates "should not necessarily be interpreted as causal"; an independent commentary concluded that the limitations prevent a conclusion that tovorafenib is inferior. The safety profiles also differ: reductions in growth velocity have been observed only with tovorafenib, whereas no ocular toxicity, cardiac dysfunction, or abnormal weight gain was observed in FIREFLY-1, and tovorafenib is dosed once weekly with a liquid formulation available and without a food effect.

Guideline positions on first-line targeted therapy vary. European standard clinical practice recommendations (version dated 2022) state that first-line targeted therapy is acceptable only within clinical trials, with possible discussion for BRAF V600E tumors. A 2024 Canadian consensus suggests upfront BRAF inhibitor with or without MEK inhibitor for most BRAF V600E gliomas for 36 months, while retaining chemotherapy as an option. A 2026 Brazilian consensus recommends considering tovorafenib for KIAA1549-BRAF fusion LGG (85% agreement) without specifying a line of therapy and recommends against type I BRAF inhibitors in that population. The 2025 NCCN pediatric CNS guideline abstract describes a scope limited to diffuse high-grade gliomas and medulloblastomas, and no NCCN recommendation for tovorafenib in pLGG was confirmed.

Key uncertainties include the following source disagreements. The FIREFLY-2 primary endpoint is described as ORR per RANO-LGG in the 2024 design publication and as ORR per RAPNO-LGG in the registry record, the 2026 sponsor announcement, and the EMA assessment, which states that the change was made after 113 of 400 participants were recruited and that consistency of the treatment effect before and after the change will be of interest. FIREFLY-1 response rates vary with the criteria and assessor (67% by IRC-assessed RANO-HGG versus 51.3% by RAPNO-LGG at June 2023, and 71.0% by IRC versus 49.3% by investigators with RANO-HGG at May 2024). For the competing first-line MEK inhibitor strategy, a 2026 review states that the selumetinib phase 3 trials (NCT03871257 and NCT04166409) closed early because of poor accrual, whereas the registry records list NCT03871257 as active, not recruiting and NCT04166409 as recruiting. Company estimates of the RAF- or BRAF-altered fraction of pLGG also differ (50% to 60% RAF-altered versus approximately 70% BRAF-altered). Comparative first-line evidence is expected from FIREFLY-2, with topline data expected by mid-2027.