CGT (Cell and Gene Therapy Access) Model

Analysis updated November 9, 2026

Strategic questions for your team

Do we have the operational and legal infrastructure to meet the June 1, 2025 State Agreement execution deadline and the Performance Period Start Date (selectable between January 1, 2025 and January 1, 2026), including an approved State Plan Amendment for value-based purchasing supplemental rebate agreements, any required state legislation or regulation changes, and documented prior authorization policies aligned with the CMS-negotiated Key Terms — and if not, what is our realistic timeline to close those gaps?
Can our billing systems and claims processing infrastructure support the 'direct reimbursement' requirement that prohibits bundling gene therapy payment into DRG or APC inpatient payments, accommodate separate NDC/HCPCS code submission for both Casgevy ($2.2M list price, Vertex) and Lyfgenia ($3.1M list price, Genetix/bluebird bio), and ensure provider reimbursement at no less than Actual Acquisition Cost as defined at 42 C.F.R. § 447.502 — and what system modifications or costs would be required before our Performance Period Start Date?
What is our financial exposure and cash flow risk across the 11-year model performance period (PY1 January 1, 2025 through PY11 December 31, 2035), given that gene therapy administration occurs in PYs 1–6, outcome measurement extends through approximately PY9, and final rebate reconciliation may not conclude until the end of the measurement period — and have we modeled the upfront drug cost liability versus the timing of outcomes-based, guaranteed, and volume-based rebate receipt from the two participating manufacturers?
Do we have — or can we contract with — at least one qualified SCD gene therapy treatment center (in-state or out-of-state per 42 C.F.R. § 431.52) that is an active CIBMTR member and can meet the patient registry participation and data submission requirements, and can we execute the required provider enrollment agreement or MOU with reimbursement rates before our Performance Period Start Date, given that starting in PY2 (January 1, 2026) a formal contract with at least one qualified center becomes mandatory?
Are we prepared to require providers to enroll in and submit data to the CIBMTR Research Database study protocol, obtain patient consent for the IRB-approved study, collect both clinical and patient-reported outcomes data over a 15-year follow-up horizon, and submit ongoing reports to CMS — and do we have the data governance, IT infrastructure, and staff capacity to fulfill these obligations for the full duration of the model, including the reconciliation period following the OBA Term?
Have we assessed our Medicaid managed care alignment obligations, specifically the requirement to amend MCO contracts and rates, issue state-directed payment preprints, ensure MCOs apply the standardized access policy identical to FFS, and guarantee continuity of care including honoring prior authorizations for at least one year post-infusion for beneficiaries who switch plans — and can we achieve full managed care alignment by our Managed Care Start Date (no later than January 1, 2026)?
Do we intend to select Casgevy, Lyfgenia, or both as State-Selected Model Drugs, and have we evaluated the Key Term variations available for preferred versus non-preferred drug designation (including the possibility of additional pricing concessions for exclusive preferred status), given that this designation decision must be disclosed in the RFA application and directly affects the magnitude of supplemental rebates and the downstream financial sustainability of covering therapies priced at $2.2M–$3.1M per patient?
Are we positioned to ensure that all enrolled Medicaid beneficiaries with SCD who are clinically eligible receive manufacturer-funded fertility preservation services at no cost to the state or beneficiary — and have we established the workflows to prevent the state from inadvertently claiming federal financial participation (FFP) for any services paid for by the manufacturer, as explicitly prohibited under Section 3.1.5 of the State RFA?
Have we applied for or evaluated eligibility for the non-competitive Cooperative Agreement funding available to participating states (with awards confirmed for CT, DC, IL, MS, NC, PA, RI, and VT), and do we have the project infrastructure to pursue implementation funding for staff and operational costs and milestone funding tied to access and care coordination performance — and what is our plan if cooperative agreement funding is unavailable or insufficient to offset participation costs?
Given the first evaluation report's finding that 84% of Medicaid beneficiaries with SCD age 12+ are already covered by participating states and that CGT participation was more than twice as high among states with the highest SCD prevalence, what is our specific eligible population size, and have we modeled the realistic number of patients likely to initiate the multi-month gene therapy care journey (including 7–9 days of chemotherapy inpatient stay, 35–45 days for infusion and recovery, and 15 years of follow-up visits) against our care management and network capacity — ensuring we neither underestimate operational burden nor overpay administrative overhead for a population that may remain small in absolute terms?

Traps and concerns

Outcomes Reconciliation Lag Creates Multi-Year Cash Flow Exposure
The CGT Access Model's OBA structure creates a reconciliation timeline that can extend up to 11 years (PY1 through PY11, January 2025–December 2035). Gene therapies administered in PY1–PY6 generate cohort-specific measurement periods beginning the year after administration, with final rebate reconciliation occurring only after the measurement period concludes. States that administer a $2.2 million Casgevy dose or $3.1 million Lyfgenia dose in PY1 (2025) must carry that financial exposure on their books for years before outcomes-based rebates are settled. State Medicaid CFOs must model long-tail contingent liabilities across multiple state fiscal cycles, budget cycles, and potential changes in state administration — all without certainty about when or whether rebates will materialize. States with limited actuarial capacity may systematically mis-budget for these contingencies, creating structural FMAP reporting distortions and federal-state reconciliation disputes.
Managed Care Alignment Deadline Triggers MCO Contract Amendment Cascade
All participating states must ensure Medicaid managed care plan policies fully align with CGT Access Model requirements — including the standardized access policy, direct reimbursement (not bundled DRG) requirements, prior authorization alignment, and continuity-of-care protections — by no later than January 1, 2026 (the Managed Care Start Date). For states with large MCO markets (e.g., California, Texas, New York, Florida, all model participants), this requires amending managed care contracts, revising capitation rates, and potentially submitting state-directed payment preprints — all within a compressed window. MCOs that fail to align risk creating a two-tiered access system where FFS beneficiaries can access gene therapy under model terms but managed care enrollees cannot, exposing the state to CMS compliance action under Section 3.1.6 and potential termination of the State Agreement (SA). The evaluation report confirmed that navigating MCO contract amendments was among the most operationally complex tasks states faced during model implementation.
Registry Participation Mandate Creates Provider Chokepoint
States must condition Medicaid payment for gene therapy — and for gene therapy administration — on the treating provider being an active member of the CIBMTR (Center for International Blood and Marrow Transplant Research) registry and obtaining patient consent for the CIBMTR Research Database Protocol. Because gene therapy for SCD is administered at a limited number of highly specialized treatment centers nationally, any CIBMTR enrollment backlog, IRB delay, consent refusal, or data submission failure at a single center can block Medicaid payment for an entire state's eligible population. States with no in-state qualified SCD gene therapy provider — a documented scenario the RFA explicitly addresses by requiring out-of-state provider contracting — face compounded risk: they must contract with out-of-state centers, enroll those providers in their Medicaid program, and verify CIBMTR registry status before a single claim can be paid. Failure on any link of this chain triggers non-payment for a therapy that costs $2.2M–$3.1M per dose.
DRG Carveout Requirement Exposes States to Hospital Billing Disputes
The CGT Access Model requires that gene therapies be reimbursed as 'direct reimbursement' and not bundled into inpatient DRG or APC payments, so that MDRP rebates — including model supplemental rebates — apply. Gene therapy for SCD involves inpatient stays of 35–45 days for infusion and 7–9 days for chemotherapy, environments where hospitals default to DRG-based billing. States must carve these drugs out of DRG bundles in their state plan and Medicaid Management Information Systems (MMIS), issue new provider billing instructions, and update claims processing business rules — all by their Performance Period Start Date. The Billing Information Guide further specifies that Casgevy must be reimbursed separately from ALL ancillary services, while Lyfgenia permits inclusive payments only if drug costs are itemized with NDC identifiers and unit counts. MMIS systems in many states are not configured for this level of drug-specific carveout for inpatient claims, and incorrect claims processing could trigger post-payment audits, claw-backs, or federal disallowances if FFP is claimed on improperly bundled payments.
Fertility Preservation Safe Harbor Boundary Creates AKS Exposure
Manufacturers participating in the CGT Access Model are required to pay for a defined scope of fertility preservation services — including travel and lodging for ancillary services — at no cost to beneficiaries or state Medicaid programs. This manufacturer-funded benefit is protected under the 'CMS-sponsored model' safe harbor at 42 CFR §1001.952(ii), which eliminates the need for a separate fraud and abuse waiver. However, the safe harbor protection is strictly bounded: it applies solely to the CGT Access Model, and only to manufacturers that meet all program requirements as outlined in the Manufacturer PA. Any manufacturer payment for fertility services that exceeds the defined scope, is made to a provider not meeting program qualifications, or is structured in a way that deviates from the Manufacturer PA terms falls outside the safe harbor and could constitute prohibited remuneration under the Anti-Kickback Statute. States must also certify that non-federal share funding does not include manufacturer-funded amounts (per Section 3.1.5), meaning any inadvertent commingling of manufacturer fertility payments with state Medicaid expenditures could trigger FFP disallowance and OIG scrutiny.
Political and Budgetary Disruption Risk Under Trump Administration
The CGT Access Model was designed and initiated under the Biden administration but is now operating under the Trump administration, with CMS led by Administrator Dr. Mehmet Oz and CMMI directed by Abe Sutton. The Trump administration's broader agenda — including DOGE-driven federal spending reductions led by Elon Musk — creates material risk that CMMI operational budgets, technical assistance resources, or the cooperative agreement funding awarded to 8 states (Connecticut, District of Columbia, Illinois, Mississippi, North Carolina, Pennsylvania, Rhode Island, and Vermont) could be reduced, rescinded, or delayed. CMS retains explicit authority under Section 2.7 of the RFA to 'modify any Model policy or parameter on an annual basis, or more frequently' and may 'modify the terms of the Model or cancel it entirely.' A model cancellation or Key Terms renegotiation mid-OBA-Term would strand states mid-measurement-period with outstanding financial obligations to manufacturers, unreconciled rebates, and beneficiaries mid-treatment-journey — a scenario the 35-45 day infusion/recovery inpatient timeline makes particularly acute.
Coverage Shift and Medicaid Disenrollment Breaks Outcome Measurement Chain
The CGT Access Model's outcomes-based rebates are triggered only if pre-specified clinical outcomes are not achieved over a multi-year Measurement Period. However, SCD disproportionately affects low-income populations with high Medicaid churn rates. A beneficiary who receives gene therapy in PY2 (2026) but loses Medicaid eligibility before the Measurement Period concludes creates a 'coverage shift' — a defined Key Term that must specify how such cases are handled in rebate calculations. The RFA explicitly lists 'Coverage Shifts' as a Key Term subject to CMS-Manufacturer negotiation, but the specific treatment of disenrolled beneficiaries' outcomes data is not publicly disclosed. If coverage shifts are excluded from outcome measurement, manufacturers face reduced rebate liability regardless of clinical failure; if included, states must maintain data-sharing relationships with beneficiaries no longer in their Medicaid program. Either approach creates financial and data governance complexity that states with limited HIT infrastructure — particularly smaller-population states that the evaluation report noted were 'less likely to participate' — will struggle to operationalize across an 11-year performance period.