
ALK lung-cancer pill and RNA for myasthenia — plus 2 stories
4 stories this week: lung cancer after an earlier targeted pill, muscle weakness in myasthenia, protein leaking into urine, and shorter tuberculosis regimens still under test. Each note says who was studied and where the result stops.
What follows is background for a general reader. A clinician who knows the person still decides whether any of it applies.
FDA grants accelerated approval to an ALK pill after earlier treatment

On 17 November 2026 the US Food and Drug Administration granted accelerated approval to neladalkib (Nelkarmi, Nuvalent) for adults with locally advanced or metastatic ALK-positive non-small cell lung cancer. The earlier treatment, given when the cancer was already unresectable and locally advanced or metastatic, had to be either one ALK tyrosine kinase inhibitor consisting of alectinib, or two or more ALK tyrosine kinase inhibitors plus platinum-based chemotherapy.
The figures come from ALKOVE-1 (NCT05384626), a multicenter single-arm open-label trial that the registry lists as phase 1/2. Among 67 people whose one prior ALK inhibitor was alectinib, the confirmed response rate by blinded central review was 49% (95% CI 37 to 62). Of those who responded, 76% still had a response at 6 months and 52% at 12 months. Among 155 people who had received two or more prior ALK inhibitors and platinum chemotherapy, the confirmed response rate was 32% (95% CI 24 to 40); 69% of responders had a response lasting at least 6 months and 51% at least 12 months. Response and its length were judged by RECIST version 1.1.
These subsets have no randomized comparison, so the rates do not show that neladalkib lengthens life against another drug. Accelerated approval is not regular approval: it uses a response rate the agency judged reasonably likely to predict benefit, and it can be withdrawn if later trials do not confirm that benefit. The label warns of liver, kidney and eye toxicity, a rise in creatine phosphokinase, and harm to a fetus. Adverse reactions reported in at least 20% of patients were liver toxicity, musculoskeletal pain, constipation, an altered sense of taste, peripheral neuropathy, swelling and fatigue.
FDA approves an RNA medicine for generalized myasthenia gravis

On 18 November 2026 Regeneron announced that the FDA had approved cemdisiran (Livsayis) for adults with generalized myasthenia gravis who have antibodies against the acetylcholine receptor, the common form of the disease. Cemdisiran is a small interfering RNA directed at complement protein C5. The company calls it the first approved siRNA for this disease and the only FDA-approved treatment for generalized myasthenia gravis given under the skin four times a year.
The approval uses the phase 3 NIMBLE trial (NCT05070858). The company release gives a least-squares mean improvement on the Myasthenia Gravis Activities of Daily Living score of 4.5 points at 24 weeks (standard error 0.4) with Livsayis, against 2.2 points with placebo (standard error 0.5). The placebo-adjusted difference was 2.3 points (p=0.0005). A gain of at least 3 points on that score occurred in 76.6% on Livsayis and 44.1% on placebo (p<0.001). On the clinician-scored Quantitative Myasthenia Gravis scale the improvement was 4.2 points (standard error 0.6) versus 1.5 (standard error 0.7), a placebo-adjusted 2.8 points (p=0.0015). A gain of at least 5 points on that scale occurred in 48.4% versus 19.0% (p<0.001). These are points on symptom scales. The approval release does not state how many people were in the analysis set behind those percentages.
The same release says the prescribing information carries a boxed warning for serious meningococcal infections, consistent with other medicines that lower C5 activity, and warnings about other serious infections. Meningococcal vaccines are to be completed at least two weeks before the first dose; if treatment cannot wait, vaccines are to be given as soon as possible and antibiotics are also required. Adverse reactions in at least 5% of people on Livsayis, and more often than on placebo, were injection-site reactions, viral infections and urinary tract infections. ClinicalTrials.gov lists NIMBLE as a randomized phase 3 study that is active and not recruiting, with 288 people enrolled. That registry count is enrollment, not the analysis set for the percentages above.
FDA approves obinutuzumab for primary membranous nephropathy

On 16 November 2026 Roche announced FDA approval of obinutuzumab (Gazyva in the United States) for adults with primary membranous nephropathy who are at risk of progression, to induce a remission of proteinuria. Roche describes it as the first therapy the FDA has approved specifically for this disease. The antibody already had other kidney uses; this note is only about the new one.
The trial is MAJESTY (NCT04629248), a phase 3 randomized open-label study. Roche says it enrolled 142 adults, assigned 1:1 to obinutuzumab or tacrolimus, which matches the registry enrollment of 142. Complete remission of proteinuria at 104 weeks was 36.1% with obinutuzumab versus 5.7% with tacrolimus, an adjusted difference of 31.1 percentage points (95% CI 18.2 to 44.0, p<0.001). Complete or partial remission at week 104 was 50% versus 12.9%, an adjusted difference of 37.8 points (95% CI 23.9 to 51.8, p<0.001). Complete remission at week 76 was 36.1% versus 8.6%, an adjusted difference of 27.4 points (95% CI 13.4 to 39.8, p<0.001).
The same trial in the New England Journal of Medicine (Fervenza and colleagues, PMID 42246654) counted complete remission at week 104 in 26 of 71 people versus 4 of 70, reported as 37% versus 6% after multiple imputation, with an adjusted difference of 31 percentage points (95% CI 18 to 44). That is the same comparison written with counts rather than Roche's adjusted percentages, not a second trial. The study is open-label, and the comparator is tacrolimus rather than another antibody. The endpoint is remission of protein in the urine with a stable filtration rate, not a demonstration that kidney failure was prevented. Roche says the safety findings matched the known profile and that no new safety signal appeared. The journal report says one person in each group died during escape therapy.
A tuberculosis platform trial flags two shorter regimens for further study

On 19 November 2026 the New England Journal of Medicine published a phase 2 randomized trial, PARADIGM4TB (NCT06114628, PMID 43056221), in people with rifampin-susceptible pulmonary tuberculosis. The registry title calls the same study a seamless phase 2B/C platform. People were assigned to standard isoniazid, rifampin, pyrazinamide and ethambutol for 24 weeks; or to bedaquiline, delamanid and moxifloxacin for 16 weeks; or to one of ten new four-drug combinations for 16 weeks. The new combinations contained ganfeborole, BTZ-043, or both, together with drugs drawn from bedaquiline, delamanid, moxifloxacin, pyrazinamide, pretomanid and linezolid.
The primary measure was the slope of log10 time to positivity of a sputum culture from baseline to week 12. That is a microbiologic signal, not a count of people cured. A Bayesian posterior probability that a new regimen beat standard treatment was used to select regimens for further study, not to declare that a regimen works. Of 401 people randomized, 371 entered the modified intention-to-treat analysis and 400 the safety analysis. Two regimens were flagged. Ganfeborole, bedaquiline, pretomanid and moxifloxacin had a slope of 0.129 log10(days) per week (95% credible interval 0.112 to 0.146), against 0.118 (0.105 to 0.131) for standard treatment: a relative difference of 9.1% (credible interval −8.6 to 29.0) and a posterior probability of superiority of 0.83. BTZ-043, bedaquiline, delamanid and moxifloxacin had a slope of 0.142 (0.126 to 0.160), a relative difference of 20.5% (2.7 to 41.1) and a posterior probability of 0.99.
Adverse events of grade 3 or higher occurred in 38% on the ganfeborole combination, and none of those people stopped for that reason; in 28% on the BTZ-043 combination, of whom one stopped because of drug reaction with eosinophilia and systemic symptoms; and in 24% on standard treatment, of whom none stopped. A separate combination of bedaquiline, moxifloxacin, pyrazinamide and BTZ-043 was discontinued because of liver toxicity. The authors say the two flagged regimens showed evidence of potential efficacy and a safety profile that did not preclude further evaluation. That is a selection step, not proof of a shorter cure.
Sources
FDA, 17 November 2026. Accelerated approval of neladalkib (Nelkarmi). https://www.fda.gov/drugs/cancer-approvals/fda-grants-accelerated-approval-neladalkib-previously-treated-locally-advanced-or-metastatic-alk
Regeneron, 18 November 2026. Livsayis (cemdisiran) approval release. https://investor.regeneron.com/node/32661/pdf
The same Regeneron release on GlobeNewswire. https://www.globenewswire.com/news-release/2026/11/18/3406224/0/en/livsayis-cemdisiran-approved-by-the-fda-as-first-and-only-sirna-therapy-in-generalized-myasthenia-gravis-and-only-treatment-with-four-times-a-year-subcutaneous-dosing.html
Reuters, 18 November 2026. Report of the Livsayis approval, including the boxed warning. https://www.reuters.com/business/healthcare-pharmaceuticals/us-fda-approves-regenerons-drug-generalized-myasthenia-gravis-2026-11-18/
Roche, 16 November 2026. FDA approval of Gazyva for primary membranous nephropathy. https://www.roche.com/media/releases/med-cor-2026-11-16
Fervenza FC et al. Obinutuzumab or tacrolimus in primary membranous nephropathy. N Engl J Med. PMID 42246654. https://pubmed.ncbi.nlm.nih.gov/42246654/
Esmail H et al. Assessing multiple treatment regimens for tuberculosis. N Engl J Med. 19 November 2026. PMID 43056221. https://pubmed.ncbi.nlm.nih.gov/43056221/
ClinicalTrials.gov. ALKOVE-1 NCT05384626, NIMBLE NCT05070858, MAJESTY NCT04629248, PARADIGM4TB NCT06114628. https://clinicaltrials.gov/
Articles in this section are educational and are not medical advice, a diagnosis, or a prescription. Consult a qualified professional before acting on anything you read here.
Read next
ResearchBiomarkersThe artery-age score was age and pressure in disguise
A person past sixty who is handed a second age can see which kind of number still sorts deaths, and which kind only repeats the page it was built from. Nothing here is a reason to chase a lower score.
ResearchMedicine and treatmentNiemann-Pick, Gaucher nerves, childhood BMI: four stories
4 stories this week: a rare childhood storage disease, neurological symptoms of Gaucher disease, breast cancer with a receptor mutation, and excess weight in children under 12. Each note says who was studied and where the result stops.
Digest · 4 stories
ResearchMedicine and treatmentA Duchenne cell therapy was refused over unreliable data
For families following Duchenne research, a refusal can say more than a promise. Here is how to read one: what the regulator set aside, and what that decision leaves open.