NIH grant cancellation legality rests on 2 CFR 200.340: a grant may be terminated mid-cycle only if the recipient fails to comply with award terms, both parties consent, or the awarding agency determines the project no longer effectuates programme goals — and in every case NIH must issue written notice and preserve the recipient’s appeal rights before funding stops.
A grant termination is the enforceable act of ending some or all of an active federal award before its approved project period expires, distinct from the routine non-renewal of a grant at the end of a competitive cycle. Understanding where that line sits — and what procedural protections apply on either side of it — is now a core competency for research administrators, not a hypothetical.
- What legal standard governs NIH grant termination?
- What grounds allow NIH to cancel a grant mid-cycle?
- What notice and appeal rights does 2 CFR 200 guarantee?
- How did 2025–26 litigation test these limits?
- Answer-first Q&A: what administrators are asking
- What should research administrators do now?
What legal standard governs NIH grant termination?
Federal grant terminations are governed by the OMB Uniform Guidance codified at 2 CFR 200.340, which HHS incorporates into its own grants regulations and which NIH restates in Section 8.5.2 of the NIH Grants Policy Statement (“Suspension, Termination, and Withholding of Support”). This is an administrative-law standard, not a discretionary one: an awarding agency cannot terminate a grant for an unlisted reason, however compelling it finds that reason.
NIH’s default posture, per its own policy statement, is to suspend a grant and give the recipient an opportunity for corrective action before proceeding to full termination — except where a serious deficiency or risk to health or safety justifies immediate termination.
What grounds allow NIH to cancel a grant mid-cycle?
2 CFR 200.340 recognises a closed, not open-ended, list of termination grounds. The Center for Science in the Public Interest’s litigation summary of APHA v. NIH describes these as “three limited circumstances” under the regulation.
| Ground | Regulatory basis | Typical trigger |
|---|---|---|
| Recipient non-compliance | 2 CFR 200.340(a)(1) | Failure to meet award terms and conditions, financial mismanagement, or research misconduct findings |
| Mutual agreement | 2 CFR 200.340(a)(2) | Recipient and NIH jointly agree the project is no longer viable (e.g. PI departure) |
| Agency priorities / “for cause” | 2 CFR 200.340(a)(4) | NIH determines the award “no longer effectuates the program goals or agency priorities” |
The third ground is the one currently under judicial scrutiny. It gives the agency real latitude to align funding with shifting priorities, but that latitude is not unlimited: agency reasoning must still satisfy the Administrative Procedure Act’s bar on “arbitrary and capricious” action, meaning NIH must show a reasoned, non-conclusory basis tied to the individual award rather than a blanket, category-wide directive.
What notice and appeal rights does 2 CFR 200 guarantee?
Termination is not self-executing. 2 CFR 200.341 requires NIH to provide written notice specifying the reason for termination, the effective date, and whether the termination is full or partial. Where non-compliance is the stated ground, the notice must also disclose that the termination will be reported in the federal System for Award Management (SAM.gov), a public record that can affect an institution’s future funding eligibility.
Recipients then have a two-tier route to challenge the decision:
| Stage | Forum | Typical deadline | Scope of review |
|---|---|---|---|
| First-level appeal | NIH official named in the termination notice | Per notice instructions | Procedural and factual objections to the stated grounds |
| Formal appeal | HHS Departmental Appeals Board | 30 days from final NIH decision | Whether NIH followed its own regulations and notice requirements |
Disagreement over scientific merit is generally not a valid appeal ground; DAB review focuses on whether NIH complied with its own procedural rules, not on re-litigating peer review.
How did 2025–26 litigation test these limits?
The clearest real-world stress test of this framework is American Public Health Association v. NIH (D. Mass., No. 1:25-cv-10787). Beginning in February 2025, NIH terminated a large volume of active grants tied to categories the administration disfavoured, without individualised, award-specific justification.
On 16 June 2025, District Judge William Young ruled the underlying directives and the resulting terminations arbitrary and capricious under the APA, finding the stated reasons “conclusory and bereft of reasoning.” His 23 June 2025 Partial Final Judgment declared the directives and terminations “of no effect, void, illegal, set aside, and vacated.” The First Circuit unanimously denied the government’s request to stay that judgment on 18 July 2025.
On 21 August 2025, the Supreme Court issued a narrower, 5–4 emergency-docket ruling: it left the vacatur of the NIH directives in place, but held that district courts likely lack jurisdiction to order restoration of the terminated grants themselves, because such claims sound in contract and belong before the Court of Federal Claims under the Tucker Act. That split — policy directives reviewable in district court, individual grant restoration routed to a separate contract forum — is now the operative jurisdictional map for any institution challenging a termination. As of the case’s most recent public docket update, First Circuit oral argument on the merits appeal was calendared for 6 January 2026, with the outcome not yet reflected in publicly available case summaries at the time of writing.
Separately, the Government Accountability Office found in August 2025 that NIH’s cancellation of roughly 1,800 grants violated the Impoundment Control Act of 1974, which requires the executive branch to obligate congressionally appropriated funds absent a formal rescission request to Congress. Harvard’s T.H. Chan School of Public Health, tracking the terminations independently, put the broader 2025 total at roughly 2,100 grants worth approximately $9.5 billion.
Answer-first Q&A: what administrators are asking
Are NIH grant terminations illegal?
Not inherently. A termination is lawful when NIH cites one of the three grounds in 2 CFR 200.340, issues proper written notice, and grounds its reasoning in the specific award. A federal court found a 2025 wave of terminations unlawful specifically because NIH skipped individualised justification and relied on blanket, category-based directives instead.
Can the government cancel a federal grant?
Yes — federal agencies retain statutory authority to terminate grants “to the extent authorized by law,” including when an award no longer serves programme goals. That authority is bounded by the Administrative Procedure Act, the Impoundment Control Act’s restrictions on withholding appropriated funds, and the agency’s own termination regulations.
How could researchers get a cancelled NIH grant restored?
Institutions can pursue NIH’s internal appeal, then the HHS Departmental Appeals Board, or litigate under the APA in district court against the policy directive itself. Per the Supreme Court’s August 2025 ruling, restoration of the underlying funding obligation likely requires a separate contract claim at the Court of Federal Claims.
How many NIH grants have been cancelled?
Independent tracking by Harvard’s T.H. Chan School of Public Health documented roughly 2,100 grants worth approximately $9.5 billion terminated during 2025, a volume the litigation record describes as unprecedented against decades in which such terminations were “exceedingly rare.”
What should research administrators do now?
Three practical implications follow directly from the legal standard, independent of how any single case resolves:
- Preserve every termination notice in full — the stated reason under 2 CFR 200.341 determines which appeal forum and deadline apply, and a vague or category-wide reason is itself a procedural defect worth flagging.
- Track the SAM.gov disclosure trigger — a non-compliance-based termination notice generates a public record that can affect future eligibility, so institutions should confirm the correct ground was cited.
- Route restoration claims correctly — the 2025 jurisdictional split means a policy challenge and a funds-restoration claim are no longer the same lawsuit, and misfiling in the wrong forum can cost months.
The underlying legal standard has not changed: NIH still needs one of three grounds, still owes written notice, and recipients still retain a defined appeal path under 2 CFR 200. What 2025–26 litigation changed is the practical burden of proof NIH must meet to invoke the “agency priorities” ground, and the forum in which recipients must seek a remedy. Research administration offices that build both into their grant-monitoring workflow will be far better positioned than those relying on general awareness that “terminations are being challenged in court.”
Leave a Reply