Why Cell And Gene Therapy Has Not Reached More Patients — And Why Clinical Operations Is The Bridge

By Leila Cupersmith, founder and CEO, Choice ClinOps

Cell and gene therapy (CGT) has already changed oncology and rare disease care. The next challenge is translating scientific possibility into credible evidence — and translating approved therapies into financially sustainable patient access.

The Promise-Access Gap

In oncology, CAR-T therapies have changed outcomes for selected hematologic malignancies. In rare disease, gene therapies have created treatment possibilities for inherited retinal disease, hemoglobinopathies, hemophilia, spinal muscular atrophy, leukodystrophies, inherited skin disorders, rare immune disorders, and genetic hearing loss.¹^, ²

And yet, in 2026, most oncology and rare disease patients still do not have access to an approved cell or gene therapy.

The issue is not simply that science has not advanced far enough. CGT depends on a coordinated ecosystem: discovery, translational science, manufacturing, clinical trials, regulatory review, reimbursement, sites, expertise, clinical care delivery, long-term follow-up, logistics, data infrastructure, patient support, and financing.

For clinical leaders, two translation points deserve particular attention.

The first is the clinical trial system, where a scientific concept must become credible evidence.

The second is the clinical care delivery system, where an approved therapy must become real patient access.

They are connected, but they are not the same — and they are not the whole ecosystem. A clinical trial asks whether an investigational therapy can be shown to be safe, effective, clinically meaningful, manufacturable, scalable, and financially viable to develop. Clinical care asks whether an approved therapy can be delivered safely, reliably, affordably, and equitably to eligible patients outside the trial setting. That distinction matters because a therapy can fail to reach patients at multiple points: target selection, manufacturing scale-up, trial execution, regulatory review, payer evaluation, treatment center readiness, referral, logistics, patient affordability, institutional cash flow, and post-treatment follow-up.

Clinical operations does not own every part of that ecosystem. But it often becomes the connective tissue that determines whether the science, evidence, manufacturing model, and financial model can move together.

Approved And Available Are Not The Same

One source of confusion in cell and gene therapy is that “available” can mean several different things.

A therapy may be regulator-approved, but only for patients whose disease biology matches the approved mechanism, such as a specific genetic variant, antigen expression pattern, human leukocyte antigen (HLA) type, tumor marker, or molecular subtype.

It may be clinically available only if the patient meets eligibility criteria related to disease stage, prior therapy, organ function, performance status, immune status, vector antibody status, collection feasibility, conditioning tolerance, or long-term follow-up capacity.

It may be operationally available only at select certified centers, within manufacturing capacity, through reliable chain-of-identity and chain-of-custody processes, and with coordinated treatment timing.

It may be financially available only if reimbursed by a payer, national health system, Medicaid program, employer plan, or specialty coverage pathway.

It may be institutionally available only if the treating hospital can absorb procurement, staffing, monitoring, billing, and reimbursement timing risk.

And it may be investigationally available only through a clinical trial, expanded access, compassionate use, or a country-specific managed access pathway.

Those are not interchangeable.

For patients and caregivers, this distinction is practical. A therapy can be scientifically relevant but genetically mismatched. It can be approved but not reimbursed. It can be reimbursed but geographically unreachable. It can be available at a center but not feasible because of disease progression, manufacturing timing, payer authorization delays, travel burden, or caregiver capacity.

For sponsors, the distinction is equally important. A successful trial does not automatically create a scalable treatment model. The operational and financial assumptions made during development can either support or limit post-approval access and real-world use.

The Centers for Medicare & Medicaid Services (CMS) describes the access problem directly: Cell and gene therapies may transform lives, but access remains difficult because these therapies can cost millions of dollars. Its Cell and Gene Therapy Access Model is testing outcomes-based agreements for sickle cell disease gene therapies, tying payment to outcomes and lowering costs for participating states.³

Why Approved Therapies Remain Concentrated

Approved cell and gene therapies remain concentrated in a limited number of diseases.

In oncology, the strongest cell therapy success has been in hematologic malignancies. Blood cancers are more accessible to current cell therapy platforms because engineered cells can circulate and encounter malignant cells. Solid tumors create a different challenge: tumor heterogeneity, immune suppression, antigen loss, trafficking barriers, hypoxia, and on-target/off-tumor toxicity.

In rare disease, many conditions are genetically defined, which makes them attractive for gene therapy. But identifying the causal gene is only the beginning. Developers still need a delivery approach that can reach the right tissue, achieve sufficient expression or editing, manage immune risk, support durability, and generate meaningful evidence in very small populations.

That scientific complexity is real. But it is not the whole story.

Even when the biology is promising, development can slow because the clinical trial model is difficult to execute and expensive to sustain. Small patient populations are geographically dispersed. Many patients are children. Natural history data may be incomplete. Endpoints may be novel. Specialized sites may be few. Families may need to travel. Follow-up may continue for years.

The science may be advanced. The operating model may still be fragile. The financial model may still be unresolved.

That is why investors continue to place large bets on next-generation platforms that may simplify delivery. In 2026, Reuters reported that Lilly agreed to acquire Orna Therapeutics for up to $2.4 billion to access technology intended to generate cell therapy inside the patient’s body and later reported that Lilly would acquire Kelonia Therapeutics for up to $7 billion to strengthen its cancer and in vivo CAR-T position.⁴^, ⁵

The Clinical Trial Execution And Cost Bottleneck

For investigational cell and gene therapies, a clinical trial tests the product and the entire development system.

A traditional oncology or rare disease trial may already be difficult to activate and execute. A CGT trial adds specialized site capabilities, narrow eligibility criteria, referral network dependency, collection or biopsy procedures, investigational product manufacturing, chain of identity, chain of custody, conditioning regimens, infusion readiness, acute toxicity management, and long-term follow-up.

Those activities are not clinical care delivery in the usual sense. They are part of clinical trial execution. And they also carry direct financial consequences.

Every delayed site, failed screen, manufacturing delay, missed dosing window, protocol amendment, unmanaged vendor handoff, and long-term follow-up gap increases sponsor cost and weakens development efficiency. For smaller and midsize sponsors, these issues can affect cash runway, financing milestones, investor confidence, partnership leverage, and board-level credibility.

In CGT, operational design is evidence-generation design as well as cost-discipline design. That means feasibility must go beyond asking whether an investigator is interested. Sponsors need to know whether a site can identify the right patients, execute collection, conditioning, infusion, toxicity monitoring, manufacturing coordination, caregiver support, and long-term data capture — within the sponsor’s budget, timeline, and financing assumptions.

Those questions belong early in development not after the protocol is finalized.

The Approved-Care Delivery And Reimbursement Bottleneck

An approved CGT creates a different operational and financial challenge. Once a therapy is approved, the question shifts from evidence generation to safe, reliable, and financially viable delivery in clinical practice. That requires treatment center qualification, clinician education, referral pathways, product ordering, payer authorization, manufacturing slot coordination, patient scheduling, caregiver planning, toxicity preparedness, billing workflows, reimbursement follow-up, and long-term monitoring.

This is not the same as running a clinical trial. The goals, accountability structure, data requirements, patient expectations, and financial risks are different.

For treating institutions, CGTs can create substantial cash flow exposure. A 2026 article in Transplantation and Cellular Therapy concluded that reimbursement delays after CAR-T therapy pose significant financial risk for hospitals and may cause smaller hospitals to limit monthly CAR-T volume or wait for reimbursement before treating additional patients.⁶

That reality matters for access. A center may be clinically capable but financially hesitant. A hospital may be certified but unable to tolerate reimbursement delays. A payer may cover treatment but require documentation that slows therapy. A patient may be eligible but face travel, lodging, time away from work, childcare, or caregiver costs that are not captured in the product price.

This is why access is not just a scientific issue or a reimbursement issue but an operational and financial design one.

Manufacturing Connects Both Worlds

Manufacturing is one of the clearest places where clinical trials and clinical care overlap but remain distinct. In a trial, manufacturing must support investigational treatment, protocol timing, release testing, safety monitoring, and data interpretation. In approved care, manufacturing must support reliable commercial delivery, capacity planning, turnaround time, quality consistency, payer expectations, and patient access.

Manufacturing also drives the cost of goods, capital requirements, facility strategy, a CDMO dependency, staffing needs, batch failure risk, and commercial margin. If manufacturing is unpredictable, both trials and clinical care suffer; trial evidence generation slows and approved care access narrows. The takeaway for sponsors is straightforward: Manufacturing strategy is not separate from clinical strategy.

Where AI And LLMs Can Help

AI and LLMs should be discussed through this operational and financial distinction. In research and development, AI may help with target discovery, genome editor design, base editor optimization, cell engineering, biomarker selection, and patient stratification. In 2025, researchers reported in Nature that protein language models could generate diverse CRISPR-Cas proteins and functional genome editors in human cells, including OpenCRISPR-1.⁷ A 2025 Nature Communications paper also described language model-assisted design of a more precise and compact adenine base editor.⁸

In clinical trials, AI may help analyze protocol complexity, identify referral patterns, model site feasibility, forecast recruitment, predict screen failure, support prescreening, monitor safety narratives, flag data inconsistencies, and summarize long-term follow-up trends.

In approved clinical care, AI may help support referral education, patient pathway mapping, operational scheduling, post-treatment monitoring, real-world data collection, and long-term outcomes tracking.

From a financial perspective, AI may eventually help sponsors and treatment centers identify avoidable friction earlier: Sites unlikely to activate on time, referral pathways unlikely to produce eligible patients, manufacturing slots at risk of mismatch, underperforming screening models, and long-term follow-up designs likely to become costly or incomplete.

But AI does not erase accountability. A 2026 review in Journal of Hematology & Oncology emphasized both the potential of generative AI in precision oncology and the importance of standardization, safety, and governance.⁹

AI can identify patterns, but it cannot decide whether a site is truly ready. AI can forecast operational risk, but it cannot replace sponsor governance. AI can support financial modeling, but it cannot make an unsustainable delivery model sustainable by itself.

The Leadership Imperative

CGT development requires one operating model for investigational trials and another for approved therapy delivery. Both must be designed intentionally. Both must also be financially realistic.

Sponsors need to think about and proactively plan for the trial pathway, post-approval access pathway, and cost pathway earlier and in parallel. That means asking: What evidence do regulators need? What evidence will payers need? What will sites need to execute? What will patients and caregivers need to participate? What manufacturing assumptions could become access barriers later? What reimbursement assumptions could limit center participation? What operational failures could create avoidable burn? What long-term follow-up model is realistic and fundable? What AI-enabled tools can reduce friction without replacing human judgment? Which stakeholders need to be included in discussions and decisions?

The next phase of cell and gene therapy will not be won by science alone but by teams that understand the difference between developing a therapy, testing a therapy, approving a therapy, financing a therapy, and delivering a therapy — that is accessible, equitable, and affordable in the real world.

Clinical operations sits across that continuum. It translates scientific possibility into executable trials. It helps translate trial evidence into regulatory and payer confidence. It helps protect sponsor capital by reducing avoidable operational waste. And it helps ensure that approved therapies can move beyond narrow centers of excellence into real patient access.

That is the bridge. In 2026, it may be one of the most important bridges the field still needs to build and support.

References:

1. U.S. Food and Drug Administration. “Approved Cellular and Gene Therapy Products.” FDA, 2026. Accessed 24 Apr. 2026. https://www.fda.gov/vaccines-blood-biologics/cellular-gene-therapy-products/approved-cellular-and-gene-therapy-products
2. U.S. Food and Drug Administration. “FDA Approves First-Ever Gene Therapy for Treatment of Genetic Hearing Loss under National Priority Voucher.” FDA, 23 Apr. 2026. Accessed 24 Apr. 2026. https://www.fda.gov/news-events/press-announcements/fda-approves-first-ever-gene-therapy-treatment-genetic-hearing-loss-under-national-priority-voucher
3. Centers for Medicare & Medicaid Services. “Cell and Gene Therapy (CGT) Access Model.” CMS Innovation Center, updated 11 Mar. 2026. Accessed 24 Apr. 2026. https://www.cms.gov/priorities/innovation/innovation-models/cgt
4. Ananthan, Padmanabhan, and Sriparna Roy. “Lilly Bets on Next-Generation Cell Therapy with $2.4 Billion Deal for Orna.” Reuters, 9 Feb. 2026. Accessed 24 Apr. 2026.
   https://www.reuters.com/legal/litigation/lilly-buy-orna-therapeutics-up-24-billion-2026-02-09/
5. Sharma, Mihika, Siddhi Mahatole, and Gursimran Kaur. “Eli Lilly to Buy Boston’s Kelonia for up to $7 Billion in Cancer Push.” Reuters, 20 Apr. 2026. Accessed 24 Apr. 2026. https://www.reuters.com/legal/litigation/eli-lilly-advanced-talks-acquire-kelonia-therapeutics-over-2-billion-wsj-says-2026-04-19/
6. Waldrup, M., F. Chow, M. Pirkola, et al. "Delayed Reimbursement Following CAR-T Therapy: A Financial Barrier for Smaller Institutions." Transplantation and Cellular Therapy, vol. 32, no. 2, suppl., 2026, pp. S313-S314. ScienceDirect. https://www.astctjournal.org/article/S2666-6367(25)02109-8/abstract#:~:text=Reimbursement%20delays%20pose%20significant%20financial,T%20across%20all%20hospital%20settings.
7. Ruffolo, Jeffrey A., et al. “Design of Highly Functional Genome Editors by Modelling CRISPR–Cas Sequences.” Nature, 2025. https://www.nature.com/articles/s41586-025-09298-z
8. Ren, Jingxuan, et al. “Protein-Nucleic Acid Language Model-Assisted Design of Precise and Compact Adenine Base Editor.” Nature Communications, vol. 16, article 11207, 2025, doi:10.1038/s41467-025-65311-z. https://www.nature.com/articles/s41467-025-65311-z
9. Hamamoto, Ryuji, et al. “Implementing Generative Artificial Intelligence in Precision Oncology: Safety, Governance, and Significance.” Journal of Hematology & Oncology, vol. 19, no. 1, 2026, article 14. https://link.springer.com/article/10.1186/s13045-026-01781-y

About The Author:

Leila Cupersmith is the founder and CEO of Choice ClinOps, a sponsor-side fractional and embedded clinical operations command center for small and mid-size oncology, rare disease, and cell and gene therapy sponsors. Choice ClinOps is not a CRO; it works inside the sponsor’s operating model as a senior execution-control layer across CROs, vendors, laboratories, sites, and internal teams. With more than 20 years of global Phase 1–3 clinical operations experience, Leila and her team help sponsors rescue delayed studies, strengthen governance, re-anchor external partners, and protect continuity through M&A, licensing, and CRO transitions.