Autologous NK cells come from the patient; allogeneic NK cells come from a donor. If a study record says 'allogeneic NK cells,' it has identified a donor source, not shown that the product is better than an autologous product, interchangeable with another allogeneic product, or approved. A patient's own expanded NK cells are autologous; cord-blood-derived NK cells given to a patient who did not supply them are allogeneic. In research, that source label is the first item to match when comparing protocols, not the final answer.
NK cells, or natural killer cells, are immune cells that can attack infected or tumor cells without the antigen-specific priming that T cells need. Autologous and allogeneic describe where the cells start.
Donor source is a starting point, not a scorecard
Swipe or scroll to read the full table →
| Question | Autologous NK cells | Allogeneic NK cells |
|---|---|---|
| Where do the cells start? | From the patient | From another person |
| What does the label establish? | The cell source | The cell source |
| What still has to match before studies can be compared? | Product, disease, phase, endpoint | Product, donor type, disease, phase, endpoint |
That table is deliberately narrow. It separates a definition from a judgment. In general, autologous products depend on collecting and manufacturing cells from the patient, while allogeneic products depend on a donor source and donor-related questions such as selection and compatibility. Those are practical research differences, not evidence that one source is safer or more effective. A like-for-like comparison would need the same product, disease, study phase and endpoint; these records do not provide that.
A Phase I allogeneic record asks a safety question first
ClinicalTrials.gov record NCT07225439 describes a Phase I study of rituximab plus tafasitamab with allogeneic NK cells in relapsed or refractory B-cell non-Hodgkin lymphoma. The record lists dose-limiting toxicity as the primary endpoint. It also states that the NK cells are investigational and not FDA-approved. Those details are specific to that drug combination, that cancer population and that US regulatory statement. The record is not a general approval of allogeneic NK cells, and a Phase I design is not a result: dose-limiting toxicity is a safety measure of early side effects severe enough to limit dose escalation, not proof that the regimen helps patients. The practical inference is that this record can support questions about safety and dosing for that protocol; it cannot support a claim that allogeneic NK cells outperform autologous NK cells.
A small early CAR-NK study is not a head-to-head comparison
The 2020 early study titled 'Use of CAR-Transduced Natural Killer Cells in CD19-Positive Lymphoid Tumors' reported preliminary response and safety data in 11 patients with relapsed or refractory CD19-positive lymphoid tumors. CAR means chimeric antigen receptor, an engineered receptor; transduced means the cells were genetically modified to carry it. The study was small and non-confirmatory. It is relevant because it shows a specific engineered NK product being studied in patients, but its scope is not a controlled comparison of autologous and allogeneic NK cells, and its findings cannot be generalized to all CAR-NK products. A reader should not turn an early 11-patient study into a claim that one donor source is superior.
A cord-blood Phase I/II record does not prove efficacy
The US National Cancer Institute trial record for CAR.70/IL15-transduced cord blood NK cells describes a Phase I/II study. Its stated research goals include safety and dosing, and the record does not establish efficacy. Cord blood is a donor source when the cells are given to a patient who did not supply them; in that framing, the product is allogeneic. That is a classification inference, not a reported clinical result. The record should not be merged with the B-cell NHL study just because both involve NK cells or donor-derived cells. NCT07225439 is a Phase I combination in a named lymphoma population; the NCI record is a Phase I/II cord-blood product record with safety and dosing goals.
Match the protocol before you compare the label
To compare autologous and allogeneic NK-cell research fairly, match the exact product, the disease, the study phase, the primary endpoint and the regulatory setting. The records above do not line up that way. NCT07225439 is a Phase I combination in relapsed or refractory B-cell non-Hodgkin lymphoma with dose-limiting toxicity as the primary endpoint. The 2020 study is a small early study of a CAR-transduced NK product in CD19-positive lymphoid tumors. The NCI record is a Phase I/II study of a cord-blood-derived engineered NK product with safety and dosing goals. Each answers a narrower question than 'which source is better?'
When a headline says allogeneic NK cells, ask: allogeneic compared with what? In which product? In which disease? At which phase? With which endpoint? Without those matches, the donor-source label is a description, not a result.
Medical-information note: This article explains research labels and study designs; it is not medical advice, and a person's options depend on the specific disease, protocol, eligibility criteria and clinical assessment.