GUARDrails or overreach? An analysis of 159 comment letters on the GUARD demonstration model

Our expert review of 159 unique submissions to the CMS Guarding US Medicare Against Rising Drug Costs (GUARD) model docket.

Background

CMS proposed the Guarding US Medicare Against Rising Drug Costs (GUARD) model as a mandatory Innovation Center demonstration project that would test an alternative method for calculating certain Medicare Part D inflation rebates. CMS states on the official GUARD model page that comments on the notice of proposed rulemaking were due February 23, 2026, and that the model, as proposed, would begin January 1, 2027 and run through December 31, 2031, with rebate invoicing and reconciliation extending into 2033. As of the date of this analysis, the public record still reflects a proposed model rather than a finalized one.

We review 159 unique comment letters and attachments submitted on the GUARD docket. The analysis focuses on who commented, the overall tenor of the file, the most common substantive objections, the major implementation concerns, and the leading requested modifications to the model.

Timing note: CMS has not publicly indicated when it will respond to the comments. Given the proposed January 1, 2027 launch date, market participants would likely expect final action well before year-end. By practical convention, late third quarter would be a plausible target, although that expectation is a hypothesis rather than announced by CMS.

Composition of the comment file

As is likely no surprise, the largest block of commenters came from the pharmaceutical and biotech industry, with a total of 53 of the 159 submissions coming from pharmaceutical or biotech manufacturers or associated trade associations. Within the company subset, the mix was not evenly split, but it did include large pharma, midsize and emerging biotech firms, rare-disease-focused companies, and plasma or specialty manufacturers. Patient advocacy organizations were the next-largest organized stakeholder category, with 38 letters. Provider and pharmacy organizations, including associations that would have to operationalize many of GUARD’s downstream effects, contributed another 20 letters.

Taken together, the file shows that GUARD drew concentrated attention from the parties most financially exposed to the model, but it also attracted a substantial response from patient and provider stakeholders. That mix matters: the case against GUARD was not limited to manufacturer self-interest, even though manufacturers were the single largest group.

Overall tenor of the file

The overall sentiment of the GUARD file was overwhelmingly negative. The most consistently oppositional submissions came from pharmaceutical and biotech manufacturers, pharma and biotech trade associations, rare-disease coalitions, and patient advocacy groups focused on access and innovation. By contrast, the more neutral letters tended to come from stakeholders focused on implementation mechanics rather than the basic legitimacy of the model, especially provider, pharmacy, and operationally oriented organizations.

Only one clearly affirmative outlier stood out: UnitedHealth Group. Its letter stated that the company was submitting comments in support of “efforts to improve prescription drug affordability” and that UHG “supports policy measures to improve prescription drug affordability.” In other words, the letter treated GUARD as part of a broader effort to bring US drug prices closer to those in peer markets, while still requesting clarification on manufacturer participation.

That contrast is analytically useful. Common themes of the letters in opposition to the program argued that GUARD was unlawful, economically distortive, or dangerous to patient access. The more neutral letters tended to be more focused on asking for more guardrails, more technical details, and a more workable implementation design.

First central objection: GUARD exceeds CMS statutory authority

The most common high-level objection across manufacturer letters was that GUARD exceeds the Innovation Center’s authority under section 1115A of the Social Security Act. Commenters repeatedly argued that GUARD is not a bona fide “test” of an innovative payment or service-delivery model, but rather a mandatory price-control mechanism attached to the Part D inflation rebate structure. In this telling, CMS is attempting to use demonstration authority to accomplish a major substantive rewrite of drug-pricing policy without clear congressional authorization.

The legal argument took several recurring forms. First, commenters said GUARD was not truly experimental because its basic effect - higher manufacturer liability if US prices exceed foreign benchmarks - was already known. Second, they argued that the model lacked the defined population, hypothesis-testing structure, and quality-improvement rationale that section 1115A contemplates. Third, several letters framed GUARD as an impermissibly broad, effectively nationwide intervention masquerading as a pilot. The long-form legal letters from large manufacturers and counsel-led submissions especially emphasized on this point.

One especially striking feature of the file is that the legality objection was not limited to manufacturers most directly exposed to the model. A total of 11 manufacturers that had signed White House MFN agreements and were therefore likely candidates for exemption still submitted GUARD comments, and each of them challenged the legality of the model in some form. That fact weakens any suggestion that the authority argument was simply a tactical complaint from companies that expected to bear the full economic burden.

Second central objection: reduced revenues, weaker innovation, and fewer therapies

The second major objection was that GUARD would reduce manufacturer revenue, weaken the expected return on pharmaceutical R&D, and ultimately leave patients with fewer therapies. This was not only a manufacturer talking point. Patient advocacy organizations, rare-disease coalitions, and health-policy groups often made the same argument, particularly in the orphan and rare-disease context.

The common causal chain was straightforward: if GUARD imports foreign reference pricing into the Medicare Part D ecosystem, manufacturers will face lower expected revenues and more uncertainty. If revenues and certainty fall, investors and companies will reallocate capital away from high-risk therapeutic areas. If capital moves elsewhere, fewer products - especially orphan, ultra-rare, and genetically targeted therapies - will be developed or launched. The patient-access argument therefore operated on two levels at once: direct near-term access concerns for current patients, and longer-term concerns that innovation itself would slow.

The Save Rare Treatments Task Force captured this theme especially clearly when it argued that applying international reference pricing to orphan drugs “would have a chilling effect on rare disease R&D investment and would threaten patient access to life-saving treatments.” Rare Disease Company Coalition, Pharming, Takeda, Vertex, and several others made closely related arguments, often linking GUARD to prior debates about orphan-drug treatment under the Inflation Reduction Act.

Importantly, the access concern was centered on US patients. Some letters cited foreign-market launch delays or reimbursement barriers abroad, but mostly as evidence that reference-pricing systems distort incentives and eventually rebound against US innovation and US patient access. The core concern was not primarily that non-US patients might have weaker access; it was that American patients (especially those with rare diseases) would see fewer treatments, later treatments, or narrower access.

Operational burden and supply-chain implementation concerns

Even where letters were less focused on legality, many still portrayed GUARD as operationally challenging. These comments came not only from manufacturers but also from organizations that would have to administer the model in practice, such as provider associations, specialty-pharmacy groups, PBM-aligned organizations, and managed-care stakeholders.

• Administrative complexity: Multiple commenters argued that GUARD would require major system-changes to claims processing, rebate accounting, eligibility determination, auditing, and reconciliation.
• ZIP-code-based implementation: The model’s geographic structure was viewed as a practical burden because operational stakeholders might need to apply different rules to otherwise similar patients depending on residence.
• Interaction with other federal pricing programs: Commenters repeatedly pointed to overlap with manufacturer rebates, the Medicare Drug Price Negotiation Program and maximum fair price implementation, the Manufacturer Discount Program, and 340B policy.
• Specialty-pharmacy and provider workflow disruption: Several letters warned that GUARD could destabilize specialty pharmacy operations, site-of-care management, network design, and continuity of therapy for complex patients.
• Insufficient implementation detail: Plans, PBMs, and pharmacies asked basic questions about timing, data feeds, benchmarking mechanics, mid-year corrections, and how stakeholders would know who is in or out of scope.

This set of comments is important because it came from the operational middle of the supply chain. Even where those stakeholders did not fully endorse the broader manufacturer case against GUARD, they often converged on the view that the proposed model did not yet have a stable implementation architecture.

Common requested modifications

The single most common concrete modification request was to exclude orphan and rare-disease therapies from GUARD. Across all the comments, 19 letters explicitly sought an orphan or rare-disease exclusion. These requests came from patient groups, rare-disease coalitions, and manufacturers. The persistence of that theme across stakeholder groups makes it the clearest consensus modification request in the file.

A second, narrower modification theme was to exempt small and mid-sized manufacturers. Five letters explicitly requested a carve-out, exemption, or comparable protection for smaller or emerging companies. This count is narrower than the broader universe of letters complaining that GUARD would hit small biotech harder than large incumbents; Many more letters raised the burden in principle, but only a subset turned that concern into a concrete proposed carve-out.

A third recurring, but more specialized, request was to exclude products where the commenting company did not control ex-US prices because foreign rights had been out-licensed to another company. Five letters explicitly raised some version of this ex-US rights or pricing-control objection. These letters argued that GUARD should not penalize a US manufacturer for foreign prices it cannot observe, negotiate, or influence, and that the model needs a safe harbor or alternative methodology for such circumstances.

There is also a practical irony embedded in the modification requests. If CMS were to exempt manufacturers with White House MFN agreements, exclude orphan and rare-disease therapies, and carve out small or midsize innovators, the remaining scope of the model would become substantially narrower. Several commenters implicitly used that logic to argue that the proposal is trying to solve a policy problem with a model whose defensible perimeter is too small to support the structure CMS has proposed.

Conclusion

The GUARD comment file is notable both for the breadth of opposition and for the degree of alignment across otherwise different stakeholder groups. Manufacturers predictably objected, but so did patient groups, rare-disease advocates, provider-facing organizations, and policy nonprofits. The strongest substantive objections coalesced around two claims: first, that GUARD exceeds CMS’s lawful demonstration authority; and second, that international-reference-style pricing pressure would reduce revenues, suppress investment, and leave patients with fewer treatment options over time.

At the same time, the comments suggest that even a narrower final model would face difficult design questions. The most common modification requests - excluding orphan and rare-disease drugs, protecting small and emerging manufacturers, and accounting for out-licensed ex-US rights - would all materially contract the scope of GUARD. Combined with operational concerns from pharmacies, provider associations, and managed-care stakeholders, the file suggests that CMS will have to choose between a broad but vulnerable model and a narrower model whose practical reach may be far more limited than the proposal implies.