Long COVID’s Shadow Lingers

NIH-led RECOVER studies show Long COVID is a heterogeneous, persistent condition with eight distinct symptom trajectories—about 10% met Long COVID criteria at three months and 81% of those still had symptoms at one year, with 5% experiencing relentlessly high symptom burden—while the RECOVER-NEURO trial found no cognitive rehabilitation that improved objective measures. Policy and industry responses have been mixed: HHS/BARDA cut 22 mRNA projects worth nearly $500m, prompting firms like Moderna to pare back mRNA programs even as the FDA approved an mRNA COVID booster for older/high-risk adults and antivirals/repurposed drugs (including Paxlovid and GLP-1 agonists) are under study. At the same time biopharma is shifting capital into onshore manufacturing—Novartis announced a $23bn U.S. buildout and Moderna a $140m facility upgrade—signaling a strategic pivot from broad mRNA R&D toward biologics, supply‑chain resilience and capacity investments amid uncertain therapeutic returns for Long COVID and a reappraisal of platform bets.

Analysis

NIH RECOVER analyses identify Long COVID as a heterogeneous, persistent condition: the Adult Cohort (3,659 participants tracked up to 15 months) delineated eight distinct symptom trajectories, about 10% met Long COVID criteria at three months and 81% of those still reported symptoms at one year, including 5% with a persistent high symptom burden, 12% intermittently high symptoms, and 14% who worsened after initial recovery. These subtype proportions and the association of persistent symptoms with female sex and prior hospitalization underscore a need for stratified trial design and targeted resource allocation.

The RECOVER-NEURO Phase 2, 22-site trial tested three cognitive rehabilitation approaches (adaptive online training, structured rehab, home-based brain stimulation) across five arms and found no objective improvement on the primary modified cognition scale despite subjective benefits; this null finding highlights the absence of validated pharmacological or behavioral therapies for cognitive Long COVID and raises the bar for future trials. The complexity of endpoints and heterogeneous patient courses will likely increase development costs and reduce near-term probability of clear therapeutic wins.

Policy and industry are recalibrating: HHS/BARDA in August 2025 terminated 22 mRNA projects worth nearly $500m, triggering firms such as Moderna to cut non-COVID mRNA programs and other clinical work, even as the FDA approved a new mRNA COVID booster for older/high-risk groups and companies shift capital into U.S. manufacturing (Novartis $23bn five-year plan with a 700,000-sq-ft hub; Moderna $140m upgrade). The combined effect is downward pressure on pure-play mRNA R&D return expectations but a potential opportunity set for onshore biomanufacturing and CMOs, while regulatory support for targeted boosters preserves tactical value for mRNA platforms.

Key Decisions for Investors

• Reassess and potentially underweight exposure to pure-play mRNA developers given BARDA de-scopes and Moderna's program cuts, monitor cash runway and any further R&D pruning
• Favor companies exposed to onshored manufacturing and capacity expansion (e.g., large-cap biopharma investing in U.S. plants and contract manufacturers) as beneficiaries of the $23bn/$140m investments and supply-chain resilience trends
• Monitor upcoming clinical readouts and trial designs for Long COVID therapeutics as binary catalysts—lack of objective endpoints increases downside risk, so use trial results to time exposure
• Watch policy signals from HHS/BARDA and FDA guidance on platform funding and booster approvals closely, and consider hedging or reducing cyclical biotech exposure until clearer therapeutic efficacy or sustained funding resumes