Genentech stopped its inhaled TMEM16A potentiator in patients before the mucus endpoints could read out

What was tried

Genentech ran NCT06603246 as a Phase Ic, open-label, multicenter study of inhaled GDC-6988 in adults with muco-obstructive disease. It enrolled two populations, non-cystic fibrosis bronchiectasis and chronic obstructive pulmonary disease, with FEV1 at least 40 percent predicted and stable disease and treatment for at least 28 days. Allocation was non-randomised, the model sequential, and there was no masking. GDC-6988, also known as RO7506811, was given twice daily through a smart dry powder inhaler.

The two primary outcomes were the percentage of participants with adverse events graded on the DAIDS scale and the percentage with spirometry abnormalities. The secondary outcomes carried the scientific weight: Part A compared spirometry with and without albuterol pretreatment, while Part B measured change from baseline in sputum percent solids and in whole-lung mucociliary clearance by scintigraphy, and required chronic sputum production of at least one teaspoon per day at entry.

The study started on 18 November 2024, enrolled 17 participants, and completed on 18 August 2026. The registry records it as terminated, gives the reason as "Sponsor's Decision", holds no posted results, and was last updated on 20 October 2026.

The biological hypothesis

TMEM16A, encoded by ANO1, is a calcium-activated chloride channel expressed in airway epithelium, and Open Targets flags it as small-molecule tractable.

The therapeutic idea is that potentiating TMEM16A increases epithelial anion secretion and airway surface liquid independently of CFTR, thinning mucus and restoring clearance. Preclinical work on the tool potentiator ETX001 showed increased anion secretion in cystic fibrosis human bronchial epithelial cells without altering calcium signalling, and faster tracheal mucus velocity and mucociliary clearance in ovine models both with CFTR pharmacologically blocked and with CFTR intact. That last result is the bridge from cystic fibrosis to non-CF disease, and the reason bronchiectasis and COPD became candidate indications.

Roche acquired the Enterprise Therapeutics TMEM16A potentiator portfolio in October 2020 for 75 million pounds upfront plus milestones, and Genentech took development forward. The acquisition release framed the mechanism as independent of CFTR mutational status and named non-CF muco-obstructive lung disease as a target.

The genetic anchoring is thin. Open Targets returns no direct association record for ANO1 and bronchiectasis, and for ANO1 and chronic obstructive pulmonary disease it returns a score of 0.044 carried entirely by a literature datatype score of 0.362, with no genetic evidence contributing. This program rested on channel physiology and animal mucus clearance, not on human genetics.

What actually happened

The registry reports 17 participants and a one-line sponsor reason. No efficacy, safety or biomarker data was posted, so the published Phase 1 record is the only quantitative evidence.

Two healthy-volunteer studies preceded this one. A nebulized study dosed 76 subjects from 1.5 to 150 mg, and a dry powder study dosed 41 subjects at three doses twice daily for 14 days. Adverse event rates ran from 37.5 percent at the lowest dose to 75.0 percent at both higher doses, against 57.1 percent on placebo, with no grade 3 or higher events, serious events or deaths. Day 1 plasma Cmax rose from 28.4 to 69.2 to 122 ng/mL across the three doses.

The spirometry finding is the one that matters. Without salbutamol, mean FEV1 fell about 2.5 percent at 11.2 mg and about 5.0 percent at 28 mg at 30 minutes, returning toward baseline by two hours. In the nebulized multiple-dose part, mean FEV1 fell transiently by 0.2 to 0.3 L at 45 and 75 mg twice daily, and two subjects in the 75 mg cohort withdrew after declines exceeding 15 percent on Day 6 and 20 percent on Day 12. Salbutamol pretreatment mitigated these declines, and the authors found no correlation between systemic exposure and FEV1 change. The terminated study was the first to put the compound into patients with already obstructed airways, and the first to measure whether any of this moves mucus.

openFDA returns no adverse event reports naming GDC-6988 or RO7506811, as expected for an agent that has never held a marketing authorisation, and ChEMBL holds no record under either name.

Failure mechanism, best guess

The registry supports a sponsor decision and nothing stronger. The honest classification is sponsor_decision, not efficacy failure and not safety signal, because no result from this study exists in the public record.

Three explanations fit the available facts without being established by them. The first is the therapeutic window. A potentiator that causes a transient fall in FEV1 in healthy lungs, and that needs a bronchodilator to blunt it, was being proposed for patients whose defining problem is airflow obstruction. The second is endpoint distance. Sputum percent solids and scintigraphic mucociliary clearance have never been qualified as predictors of exacerbation reduction, so even a clean Part B would have left the sponsor far from a registrational path. The third is portfolio allocation, which the registry language is consistent with and which no public document confirms.

How to prevent this next time

No endpoint-level data from this trial exists, so no posterior or power calculation can be constructed from it. The available levers are qualitative, and the Phase 1 record already pointed at them.

Enrich on mucus burden at entry rather than on diagnosis. Part B required at least one teaspoon of sputum per day, a low bar, and sputum percent solids was measured as an outcome but never used to select participants, so a trial of 17 carried unknown variance on its central biomarker.

Adjust the base rate. Inhaled agents aimed at mucus rheology in non-cystic fibrosis bronchiectasis have a thin record of converting mechanistic effect into exacerbation benefit, and the ovine clearance data came from animals with normal airway geometry.

Red-team the bronchoconstriction signal before the patient study rather than during it. Two healthy-volunteer withdrawals for FEV1 declines above 15 and 20 percent, in a program destined for obstructed patients, is a question about dose and formulation rather than a tolerability footnote to be managed with albuterol.

The single highest leverage change would have been to make sputum percent solids both an enrollment criterion and a prespecified primary endpoint in the first patient study, so that 17 participants would have produced an interpretable answer about mucus rather than a terminated safety run-in.

What this means for similar programs

TMEM16A potentiation remains a reasonable idea for cystic fibrosis, where CFTR-independent anion secretion addresses a defined molecular defect. The transfer to non-CF bronchiectasis and COPD imports the mechanism without importing the genetics, and Open Targets scores that transfer at 0.044 for COPD and records nothing for bronchiectasis. Programs that bypass a defective protein by potentiating a parallel one should expect their strongest evidence to be physiological rather than genetic, and should budget for a longer path to a clinical endpoint. ANO1 enters the Claidex graph with a single Phase 1 stop and a Mechanism Risk Score of 29, in the yellow band, of which the genetic deficit term supplies 14.34 of a possible 15 points.

Open questions

Did Part B read out, and what did sputum percent solids and mucociliary clearance show in participants who completed it? Was the FEV1 decline seen in healthy volunteers larger or dose-limiting in patients with established obstruction? Does the cystic fibrosis indication continue independently of this decision?

Sources

1. • ClinicalTrials.gov, NCT06603246, A Phase Ic, Open-label, Multicenter Study to Evaluate the Safety, Tolerability, and Activity of Inhaled GDC-6988 in Patients With Muco-obstructive Disease. https://clinicaltrials.gov/study/NCT06603246 - Miller P, Repplinger D, Zhu R, et al. Randomized, phase I studies to evaluate the safety, tolerability, and pharmacokinetics of an inhaled, TMEM16A potentiator, GDC-6988, in healthy subjects. Ther Adv Respir Dis.

2. - Zhu R, et al. Mechanistic physiologically based pharmacokinetic modeling of dry powder and nebulized formulations of orally inhaled TMEM16A potentiator GDC-6988. CPT Pharmacometrics Syst Pharmacol.

3. - Danahay HL, Lilley S, Fox R, et al. TMEM16A potentiation: a novel therapeutic approach for the treatment of cystic fibrosis. Am J Respir Crit Care Med. 2020;201(8):946-954. - Galietta LJV. TMEM16A (ANO1) as a therapeutic target in cystic fibrosis. Curr Opin Pharmacol. 2022;64:102206. - Open Targets Platform, target ANO1 (ENSG00000131620), associations with bronchiectasis (MONDO_0004822) and chronic obstructive pulmonary disease (MONDO_0005002), retrieved 23 October 2026. https://platform.opentargets.org/target/ENSG00000131620 - Enterprise Therapeutics. First-in-class TMEM16A potentiator program for treatment of cystic fibrosis and other respiratory diseases acquired by Roche. 6 October 2020. https://enterprisetherapeutics.com/enterprise-therapeutics-first-in-class-tmem16a-potentiator-program-for-treatment-of-cystic-fibrosis-and-other-respiratory-diseases-acquired-by-roche/ - openFDA drug adverse event endpoint, queries for GDC-6988 and RO7506811, retrieved 23 October 2026. https://open.fda.gov/apis/drug/event/.