Rare cancers collectively represent a significant oncology burden, creating an important development opportunity for cell and gene therapies beyond traditional blood cancers.
Recent FDA approvals of lifileucel in melanoma and afamitresgene autoleucel in synovial sarcoma show how cellular therapies are moving into solid tumor and rare oncology settings.
Emerging approaches in rare oncology include tumor-infiltrating lymphocyte therapy, TCR-engineered T cells, and investigational CAR-T strategies for difficult solid tumors.
Successful rare oncology CGT development depends on early integration of patient selection, companion diagnostics, clinical trial design, CMC strategy, and manufacturing comparability.
Small patient populations, molecular testing access, and accelerated approval confirmatory requirements remain key challenges for sponsors developing CGTs for rare cancers.
Rare Oncology as a Development Challenge
Rare oncology occupies an unusual position in cancer research and development. Each individual cancer type may affect a relatively small population, but rare cancers as a group represent a substantial share of the oncology landscape. The National Cancer Institute defines a rare cancer as one that occurs in fewer than 15 out of 100,000 people each year while also describing rare cancers as cancers that affect fewer than 40,000 people per year in the United States.1,2 Collectively, rare cancers make up just over a quarter of all cancers, and all childhood cancers are considered rare because cancer rates in children are very low.
That collective burden creates a difficult development problem. Rare cancers can be harder to prevent, diagnose, treat, and research, and the limited size of each patient population can make conventional oncology development strategies more difficult to execute.1 The U.S. Food and Drug Administration’s (FDA’s)Oncology Center of Excellence has identified several persistent challenges in rare cancer drug development, including the difficulty of enrolling enough patients in clinical trials, limited or delayed access to molecular testing, incomplete understanding of pathophysiology, molecular characteristics, and natural history, and the difficulty of conducting randomized trials in very small populations.3
These challenges are especially relevant for cell and gene therapies (CGTs), which often require a close match between tumor biology, patient selection, clinical evidence strategy, and manufacturing execution. In rare oncology, the question is rarely whether a therapeutic concept is biologically interesting in isolation. The more practical question is whether the field can identify the right patients, generate meaningful evidence in small populations, control complex manufacturing processes, and build an approach that can move from an early clinical signal to a licensable and deliverable therapy.
Moving Beyond the Hematologic Starting Point
The modern story of cellular immunotherapy has been shaped heavily by hematologic malignancies, but the field is now testing and, in select cases, advancing approaches for solid tumors and rare oncology applications. That shift should be viewed carefully. The available examples do not suggest that the barriers facing CGTs in solid tumors have been broadly solved. They do show that specific modalities, matched to specific tumor contexts, are reaching important clinical and regulatory milestones outside of the blood cancers where cellular therapy first became established.
One of the clearest milestones came with lifileucel, an autologous tumor-infiltrating lymphocyte (TIL) therapy. In February 2024, the FDA granted accelerated approval to lifileucel for adults with unresectable or metastatic melanoma after prior treatment with a programmed death receptor-1 blocking antibody and, if the tumor was BRAF V600 positive, a BRAF inhibitor with or without a MEK inhibitor.4 That approval was described as the first approval of a TIL-based cancer treatment and the first cellular therapy approved for a solid tumor.5
The melanoma example is important less because it defines rare oncology than because it expands the conversation beyond engineered chimeric antigen receptor T cell (CAR-T) therapy. TIL therapy begins from immune cells already present in the patient’s tumor rather than from a synthetic receptor design alone. It therefore represents a different model for applying cellular therapy to solid tumors and underscores that cell therapy beyond hematologic malignancies will not be defined by CAR-T alone. TIL therapy, T cell receptor (TCR)-engineered cells, and CAR-T strategies each depend on different relationships among tumor biology, target expression, patient selection, and manufacturing strategy.
That distinction matters for rare oncology. Rare cancers often require development strategies that begin with the specific biology of the disease, the feasibility of identifying eligible patients, and the evidence pathway available for small populations. The field is not simply asking whether CGTs can move beyond blood cancers. It is asking which cellular approaches can be aligned with tumor-specific biology, clinical need, and a realistic development pathway.
TCR-Engineered Cells in Synovial Sarcoma
Afamitresgene autoleucel (afami-cel) provides the most direct example of a genetically modified autologous T cell therapy reaching a rare solid tumor setting. In August 2024, the FDA granted accelerated approval to afamitresgene autoleucel for adults with unresectable or metastatic synovial sarcoma who had received prior chemotherapy and whose tumors met defined HLA-A*02 and melanoma-associated antigen A4 (MAGE-A4) expression requirements.6 The FDA described the therapy as a MAGE-A4-directed genetically modified autologous T cell immunotherapy. The agency also described Tecelra as a gene therapy for adults with metastatic synovial sarcoma meeting the specified eligibility criteria.7
This case illustrates how rare oncology CGT development can become highly specific. The therapy is not aimed at all sarcomas or all synovial sarcoma patients. Eligibility depends on prior treatment history, human leukocyte antigen (HLA) status, and tumor MAGE-A4 expression, with MAGE-A4 expression determined by FDA-approved or cleared companion diagnostic devices.6 That makes the development pathway inseparable from patient identification. In small populations, every additional selection criterion can narrow the eligible pool, but those same criteria may be essential to ensure that the therapy is being tested and used in the patients most likely to match its mechanism.
The clinical evidence behind afami-cel also reflects the realities of rare oncology development. The SPEARHEAD-1 study was an international, open-label phase II trial conducted across sites in Canada, the United States, and Europe.8 Its first cohort included patients with HLA-A02-positive, MAGE-A4-expressing metastatic or unresectable synovial sarcoma or myxoid round cell liposarcoma who had previously received an anthracycline- or ifosfamide-containing chemotherapy regimen). The abstract reports an overall response rate of 37% across the cohort, 39% in synovial sarcoma, and 25% in myxoid round cell liposarcoma. The investigators concluded that afami-cel produced durable responses in heavily pretreated patients with HLA-A02- and MAGE-A4-expressing synovial sarcoma.
The synovial sarcoma example therefore carries several lessons for rare oncology CGT development. It shows how a rare tumor can become addressable through a defined antigen and genetically modified autologous T cell strategy. It also shows how the therapeutic, diagnostic, clinical, and regulatory components of development converge. The therapy requires an eligible target, a suitable HLA context, a companion diagnostic pathway, specialized manufacturing, and a clinical development plan capable of supporting accelerated approval in a small and heavily pretreated population.
Early CAR-T Experience in Solid CNS Tumors
Beyond approved therapies, early clinical studies in central nervous system (CNS) tumors are beginning to show how cellular therapy design, delivery, and monitoring may need to be adapted for tumor settings that differ sharply from hematologic malignancies. These examples remain early clinical experiences, but they help illustrate the specialized considerations that can shape rare oncology CGT development.
One first-in-human, investigator-initiated, open-label study treated three participants with recurrent glioblastoma using intraventricular CARv3-TEAM-E T cells.9 The cells were engineered to target both EGFR variant III and wild-type EGFR through secretion of a T cell–engaging antibody molecule. In that small study, radiographic tumor regression was rapid but transient in two of three participants. The same report noted that the use of CAR-T cells in solid tumors such as glioblastoma has been limited by the challenges of single-antigen targeting in heterogeneous disease and mechanisms associated with the immunosuppressive tumor microenvironment.
That glioblastoma study is useful because it underscores the design problem facing cellular therapy in solid tumors. Antigen selection is not a simple matter of choosing a marker and building a receptor. Tumor heterogeneity can create escape routes, and the local tumor environment may constrain immune activity. The CARv3-TEAM-E strategy addressed that challenge through a design intended to target EGFRvIII and wild-type EGFR, but the early clinical experience also showed the need for caution, as the reported tumor regression was transient in the two participants who experienced it.9
A separate early example comes from H3K27M-mutated diffuse midline gliomas. A phase I trial of GD2-CAR T cells was designed to assess manufacturing feasibility, safety, tolerability, and dose selection in that setting.10 The published report presented the first four patients with H3K27M-mutated diffuse intrinsic pontine glioma or spinal cord diffuse midline glioma treated with GD2-CAR T cells, and three of the four patients exhibited clinical and radiographic improvement. The article also described the incorporation of neurocritical care precautions because CAR-T-induced brainstem inflammation can create serious risks.
These CNS studies show why rare oncology CGT development often requires highly disease-specific thinking. The same broad category, CAR-T therapy, can involve very different decisions depending on the target, the tumor site, the route of administration, the anticipated toxicities, and the monitoring needs of the patient population. For rare CNS tumors, early signals may be scientifically compelling, but the development path must account for safety risks linked to local inflammation, small patient numbers, and the need to generate evidence in populations where large randomized studies may be difficult.
Small Populations and Evidence Generation
Rare oncology development is constrained not only by biology but also by evidence generation. The FDA’s rare-cancer program identifies difficulty enrolling enough patients as one of the central challenges in rare-cancer drug development, along with limited or untimely access to molecular testing, incomplete understanding of disease pathophysiology and natural history, and difficulty conducting randomized clinical trials.3 These issues become more pronounced when a therapy is restricted to a molecularly or immunologically defined subgroup within an already rare cancer.
That problem is directly relevant to CGTs. A therapy such as afami-cel illustrates how patient selection can narrow a rare tumor population through prior treatment status, HLA type, and antigen expression.6 From a scientific perspective, such selection may be necessary to match the therapy to the intended mechanism. From a development perspective, it can make enrollment, site selection, screening, and evidence generation more complex.
The FDA’s draft guidance on innovative trial designs for CGT products in small populations provides recommendations for clinical trials of CGT products intended for diseases or conditions affecting small populations and addresses trial designs and endpoints intended to generate clinical evidence to support licensure.11 The FDA’s rare cancer program also identifies real-world data (RWD), real-world evidence (RWE), Bayesian approaches, and other statistical methodologies as relevant to rare-cancer evidence generation.3
For sponsors, these points suggest that evidence planning in rare oncology CGT programs has to begin early. The trial strategy must reflect the expected size of the eligible population, the feasibility of molecular testing, the natural history of the disease, the ethical and practical limits of randomization, and the endpoints most likely to support a meaningful regulatory decision. These considerations are not secondary to the science. In small populations, they may determine whether promising biology can be translated into interpretable clinical evidence.
Patient Selection and Diagnostic Infrastructure
Patient selection is a central feature of rare oncology CGT development, not merely a screening step. The afami-cel approval makes this especially clear. The FDA indication requires HLA-A*02 positivity and tumor MAGE-A4 expression, with MAGE-A4 expression determined by FDA-approved or cleared companion diagnostic devices.6 That means the therapy’s use depends on the ability to identify eligible patients through specific testing pathways.
In rare oncology, diagnostic infrastructure can affect development in several ways. It can determine whether patients are identified early enough for trial referral, whether enrollment targets are feasible, and whether a commercial therapy can reach the patients described in its label. The FDA’s rare cancer program identifies limited or untimely access to molecular testing as a development challenge, which is particularly relevant for therapies that depend on antigen expression, HLA type, or other biomarker-defined eligibility criteria.3
The FDA’s guidance on in vitro companion diagnostic devices describes considerations for development and labeling of companion diagnostics to support the indicated uses of multiple oncology drug or biological products when appropriate.12 For rare oncology CGTs, companion diagnostics can therefore sit at the intersection of therapeutic development, clinical trial execution, and post-approval access. A therapy may have a strong mechanistic rationale, but if eligible patients cannot be found through reliable and accessible testing, the development pathway becomes harder to sustain.
This dynamic also affects how sponsors and development partners should think about program design. A rare oncology CGT program may need to integrate therapeutic development, assay strategy, site education, referral networks, and patient-identification plans earlier than a more conventional oncology program. That does not mean every program will require the same diagnostic model. It does mean that the ability to find the right patients is part of the product-development problem from the beginning.
CMC Expectations and Manufacturing Change
Rare oncology may involve small patient populations, but CGT programs still require substantial manufacturing and quality control. For CAR-T products, key development considerations include chemistry, manufacturing, and controls (CMC), pharmacology/toxicology, clinical study design, and analytical comparability, and many of those same considerations may apply to other genetically modified lymphocyte products, including CAR natural killer cells and TCR-modified T cells.13
For human gene therapy investigational new drug applications (INDs), FDA guidance states that sponsors should provide sufficient CMC information to assure product safety, identity, quality, purity, and strength, including potency.14 That expectation is central for rare oncology CGTs because many of these products are complex, patient-derived, genetically modified, or otherwise difficult to characterize using conventional product paradigms. A small patient population may influence clinical strategy, but it does not eliminate the need to understand what is being manufactured and how product quality will be assessed.
At the same time, the FDA has recognized the need for flexibility in the development of human cellular and gene therapy products. The agency’s 2026 guidance on CMC flexibilities describes how it applies flexibility to CMC requirements for human CGT products being developed for biologics license applications (BLAs).15 That flexibility is intended to help expedite development, review, and patient access for safe and effective CGT products addressing serious or life-threatening conditions with significant unmet medical needs.
However, flexibility should not be confused with loosened expectations around product understanding. FDA guidance on manufacturing changes and comparability states that managing manufacturing changes presents challenges for human cellular therapy and gene therapy products because of product complexity.16 The guidance addresses management and reporting of manufacturing changes and comparability studies to assess the effects of changes on product quality. For rare oncology CGTs, this is a practical development issue. Processes may evolve between early clinical studies and a BLA, but those changes must be understood, justified, and connected to product quality.
This is where development partners can play an important role. A drug developer advancing a rare oncology CGT may need to move quickly because the clinical need is high and the patient population is small. At the same time, the program must generate enough CMC knowledge to support clinical testing, manufacturing changes, comparability, and eventual licensure. A strong development strategy therefore has to treat CMC not as a late-stage documentation exercise but as a core part of the translational plan.
Accelerated Approval and Confirmatory Evidence
Both lifileucel and afamitresgene autoleucel received accelerated approval, which reflects the relevance of that pathway for serious oncology settings where there is an unmet need and sufficient evidence to support the regulatory standard.4,6 The accelerated approval framework can be particularly important in rare oncology because conventional development pathways may be difficult to execute when populations are small, disease biology is heterogeneous, and randomized trials are challenging.
At the same time, accelerated approval is not the endpoint of development. The FDA’s 2025 draft guidance on accelerated approval states that the agency may require confirmatory studies to be underway before accelerated approval or within a specified time period after approval.17 For rare oncology CGTs, that expectation must be considered early because postapproval evidence generation may be difficult if the eligible population is small, geographically dispersed, or further restricted by molecular or diagnostic criteria.
The afami-cel example shows how these issues can converge. The therapy’s indication depends on prior treatment, HLA-A*02 status, and MAGE-A4 expression determined through companion diagnostic devices.6 Those same factors that define the appropriate patient population may also complicate confirmatory evidence generation. The development strategy therefore must anticipate how patients will be identified, how sites will be activated, how evidence will be collected, and how manufacturing capacity will align with clinical and regulatory commitments.
For drug developers and their partners, accelerated approval should be viewed as part of a broader evidence plan rather than a shortcut around it. In rare oncology CGT development, the confirmatory strategy, diagnostic strategy, and manufacturing strategy all need to be aligned. If any one of those pieces is delayed or underdeveloped, the program may struggle to convert an early signal into durable clinical and regulatory progress.
Implications for Biopharma and Development Partners
Rare oncology CGT programs bring together several kinds of complexity that are often discussed separately. The biology may require an antigen, immune context, or patient-derived cellular starting point that is specific to a narrow disease population. The clinical program may need to generate evidence in a small cohort where randomized trials are difficult. The diagnostic strategy may determine whether eligible patients can be found at all. The manufacturing strategy must support product identity, quality, purity, strength, and potency, while also anticipating process changes and comparability requirements.14,16
That convergence changes how biopharma companies should think about early development. For a conventional oncology program, some operational elements can often be refined as the program advances. In rare oncology CGT development, delayed planning can create larger consequences. If molecular testing access is limited, enrollment may lag. If patient-selection criteria are not aligned with the clinical protocol and diagnostic pathway, the trial may struggle to find the intended population. If manufacturing changes are not anticipated, comparability questions may slow development later. If confirmatory evidence planning begins only after accelerated approval, the program may face avoidable constraints.
The examples discussed here suggest that the most successful rare oncology CGT programs will likely be built around integration from the beginning. Lifileucel demonstrates that a patient-derived cellular therapy can reach a solid tumor approval milestone. Afami-cel shows how a genetically modified autologous T cell therapy can be developed for a rare sarcoma population defined by HLA type and antigen expression. Early CAR-T studies in glioblastoma and diffuse midline glioma show that investigators are testing increasingly specialized strategies in difficult CNS tumors, while also encountering the safety, durability, and disease-specific challenges that come with those settings.
For development partners, including contract development and manufacturing organizations (CDMOs), the opportunity is not only to manufacture a complex product. It is to help sponsors build development systems that are appropriate for small populations and advanced modalities. That may include early CMC planning, analytical strategy, process understanding, comparability planning, and coordination with the clinical and diagnostic elements of the program. The more specialized the therapy and the smaller the population, the more important it becomes to avoid treating these as disconnected workstreams.
A More Precise Future for Rare Oncology
Rare oncology is likely to remain a difficult setting for CGT development because the underlying constraints are structural. Patient populations are small, disease biology may be incompletely understood, molecular testing access can be limited, and randomized trials may be difficult to conduct. CGTs add further complexity because they require rigorous CMC strategies, product-specific comparability planning, and, in many cases, specialized patient-selection infrastructure.
Yet the recent evidence base supports cautious optimism. Lifileucel established the first TIL therapy approval and the first cellular therapy approval for a solid tumor. Afami-cel brought a MAGE-A4-directed genetically modified autologous T cell immunotherapy into an FDA-approved rare sarcoma setting with defined HLA and antigen-expression requirements. Early CAR-T studies in recurrent glioblastoma and H3K27M-mutated diffuse midline glioma show continued experimentation with cellular therapy designs for solid CNS tumors, even as they highlight the need for careful interpretation and specialized clinical management.
The path forward will not be defined by a single platform or a simple extension of hematologic cell therapy models into solid tumors. Rare oncology will require modality-specific development, precise patient identification, practical evidence strategies for small populations, and manufacturing approaches that can support both flexibility and rigorous product understanding. For sponsors, CDMOs, clinical investigators, diagnostic partners, and regulators, the central task is to build programs in which those elements are connected early enough to give promising therapies a realistic path to patients.
References
1. “Rare cancer.” National Cancer Institute. Dictionary of Cancer Terms. Accessed 29 Jun. 2026.
