September 2026 Research Update

This month added early human evidence that some newer or less commonly used medicines may reduce LGLL cells and improve blood counts for many people, at least in the short term. In two small retrospective studies, most patients responded to treatment, and responses were seen even in groups that can be hard to treat, such as relapsed or refractory disease. Because these were not randomized trials and included small numbers of patients, the main value right now is helping clarify what options look promising enough to study further. [1] [2]

In a single-center report of 16 people with relapsed or refractory LGLL, golidocitinib (a JAK1 inhibitor) was linked with rapid improvement in blood counts for most patients, with responses reported within a few weeks on average. The study included both T-LGLL and NK-LGLL, and responses were reported in both groups, but follow-up was short at about 4 months, so it could not fully address how long responses last. Side effects were common but were mostly mild, with a few grade 3 blood count or liver enzyme problems reported. [1]

A separate retrospective cohort of 12 people with high-burden T-LGLL reported that bendamustine chemotherapy produced clinical responses in all patients, with responses typically taking a few months. The report also described longer follow-up than the golidocitinib study, including a 3-year progression-free survival estimate, and noted that fewer than half had severe (grade 3 to 4) side effects, although two patients stopped treatment due to toxicity. Laboratory work done on patients' cells supported a direct effect of bendamustine on leukemic LGL cells and suggested it can reduce STAT3 signaling, and responses were seen whether or not STAT3 mutations were present. [2]

For patients and caregivers, these studies mainly add to the picture of what treatments might be available when LGLL is harder to control or has a higher disease burden. The reports suggest that some people may see blood counts improve relatively quickly with golidocitinib, while bendamustine responses may take longer but were still reported broadly in the studied group. What remains uncertain is how well these results will hold up in larger groups, how long responses last for most patients, and how these options compare head-to-head with other approaches, since both reports were small and not randomized. [1] [2]

This month also gave a clearer short-term safety snapshot: both treatments had side effects that sometimes involved low blood counts, and bendamustine led to treatment stopping for a couple of patients. These details may matter later if future studies confirm benefit and help define who is most likely to respond and what monitoring is needed during treatment. For now, the evidence is best viewed as early, real-world signals rather than final answers about the best choice for any one person. [1] [2]

This month emphasized that accurately identifying LGLL, and separating it from look-alike reactive immune cell expansions, depends heavily on consistent lab interpretation and agreed-upon diagnostic steps. One paper focused on how to read NK-cell patterns in flow testing in a more reliable way, especially when trying to judge whether an NK-cell expansion is clonal (suggesting a neoplasm) or reactive (suggesting an immune response). Another paper offered a broad expert consensus on diagnosing LGLL and following patients over time, aiming to make care more standardized within its setting. [3] [4]

The flow cytometry best-practices manuscript highlighted that some NK-cell patterns can be misleading and that clonality inference using KIR receptors (a set of NK-cell surface markers) has important pitfalls. It noted, for example, that certain NK-cell subsets may not be well served by KIR-based approaches and that reactive NK-cell expansions can sometimes look KIR-restricted, which could falsely resemble a neoplastic clone. To reduce misclassification, it proposed an operational reflex approach that is triggered by an NK-cell expansion or certain abnormal marker patterns, and it used case examples that included NK-LGL leukemia and reactive adaptive NK expansions. [3]

The Chinese expert consensus covered diagnostic criteria and differential diagnosis for LGLL, and it also summarized commonly used medicines and newer targeted drugs that have shown preliminary activity in prior reports. It described first-line immunosuppressive options such as cyclophosphamide, methotrexate, and cyclosporine A, and it discussed follow-up expectations such as reassessment after several months of therapy. The document also listed safety concerns that can matter in day-to-day life, such as infection risk with immune-suppressing treatments and drug-specific risks like liver or kidney problems. [4]

For patients, this set of papers matters most for understanding why LGLL diagnosis can take careful work, and why results sometimes depend on exactly how specialized lab tests are interpreted. The flow-testing paper helps explain how labs try to avoid confusing a true NK-cell leukemia with a temporary or reactive NK-cell rise, which can affect how confident the diagnosis is. The consensus statement does not provide new trial results, but it may matter now by showing what one national expert group considers a standard diagnostic workup and how they think about follow-up and treatment safety issues. [3] [4]

What remains uncertain is how consistently these diagnostic approaches perform across different labs and patient populations, because these publications did not report new quantitative accuracy outcomes. Still, they help clarify the direction of the field: more standardized testing language, clearer decision steps, and greater attention to situations where a lab pattern can mimic LGLL. Over time, that kind of standardization can make research studies easier to compare and can reduce confusion for patients who receive differing opinions. [3] [4]