Going once, going twice, sold - not to miracle hype, but to something much rarer in relapsed peripheral T-cell lymphoma: a result that actually makes oncologists put down their coffee and squint at the screen again.
Peripheral T-cell lymphoma, or PTCL, is one of those cancers that sounds like it was named by a committee trapped in a fluorescent conference room. In real life, it is a rough group of rare, aggressive lymphomas that start in mature T cells, the immune system’s hired security staff. Except now the security staff has gone rogue, and the building is on fire. Not ideal.
That is why the final phase 2 PRIMO trial results for duvelisib matter. This is an oral drug that blocks two PI3K signaling switches, delta and gamma, that help lymphoma cells survive and may also help shape the tumor’s weird little support network. In the PRIMO study, patients with relapsed or refractory PTCL got duvelisib at 75 mg twice daily for 2 cycles, then 25 mg twice daily to keep the pressure on while trying to limit later toxicity. The headline: 48.0% responded, and 33.3% had complete responses. In the AITL subgroup, the numbers looked even stronger: 62.2% responded and 51.4% had complete responses.[1]
Why this catches your eye
Relapsed PTCL is not a disease with an overflowing buffet of great options. It is more like opening the fridge at midnight and finding mustard, half a pickle, and regret. These lymphomas are biologically messy, clinically diverse, and stubbornly hard to treat. Even newer reviews still describe PTCL as a field held back by heterogeneity, resistance, and a shortage of clean, subtype-matched therapies.[2,3]
So when a single oral drug produces that many complete responses, people notice.
And not just because oncologists enjoy percentages. A complete response in this setting can mean time - time to feel better, time to avoid more toxic salvage therapy, time to reach transplant in selected patients, time to get your life out of pure crisis mode. Cancer biology is complicated, but the human version is not. A deeper remission can buy breathing room, and breathing room is expensive.
The sci-fi part, because modern oncology has fully left the chat
Duvelisib works on PI3K-delta and PI3K-gamma, which are part of the signaling circuitry cells use to grow, survive, and communicate. If PTCL were a dystopian space station, PI3K would be part of the power grid and the intercom system. Cut the right wires, and you do not just annoy the bad actors - you can disrupt the whole shady operation.
That may be especially relevant in AITL and related T-follicular helper-type lymphomas, where the disease often seems less like a lone villain and more like a conspiracy with excellent catering. The malignant T cells do not act alone. They manipulate surrounding immune cells, stir up the microenvironment, and turn lymph nodes into chaotic little diplomatic disasters. Reviews of PTCL biology and therapy keep pointing to this subtype-specific complexity, which is exactly why the stronger AITL signal in PRIMO feels more than random.[2,3]
Plot twist: the weirdest tumors may finally be telling us what they are vulnerable to.
Before we start engraving statues, here’s the catch
This was not a fairy tale with zero side effects. 97.6% of patients had treatment-emergent adverse events, and 74.0% had grade 3 or higher events.[1] That is serious. The trial’s dosing strategy itself reflects that reality: start harder, then step down to try to preserve benefit while reducing late toxicity.
Also, the median progression-free survival for the overall group was 3.4 months, which sounds less dramatic than the response rate because PTCL loves ruining a good party.[1] The median duration of response was 7.9 months overall and 11.3 months in AITL.[1] So this is not “problem solved.” This is “we may finally have a tool that fits part of the lock.”
That matters because PTCL treatment is moving toward combinations and subtype-specific strategies, not one-size-fits-all chemo with vibes. A 2024 Nature Medicine study of duvelisib plus romidepsin showed encouraging activity in relapsed/refractory T-cell lymphomas, especially in PTCL, which suggests PI3K blockade may be even more useful as part of a smarter combo rather than a solo act forever.[4]
Why this paper is actually fun, in the nerdiest possible way
PRIMO is interesting because it hints that PTCL may be less “one disease with bad manners” and more a cluster of biologically different enemies wearing similar uniforms. That idea has been building for years, especially as newer biology work shows just how diverse T-cell lymphomas really are.[3] If duvelisib works best in the T-follicular helper neighborhood, that is not a footnote. That is the map starting to sharpen.
And once the map sharpens, drug development stops looking like blindfolded darts and starts looking more like targeting.
Which, in PTCL, would be a nice change of pace.
References
-
Mehta-Shah N, Zinzani PL, Jacobsen ED, et al. Duvelisib Induces Deep Responses in PTCL: Final Results of the Phase 2 PRIMO Trial of Duvelisib in Relapsed/Refractory Peripheral T-Cell Lymphoma. Journal of Clinical Oncology. Published April 22, 2026. DOI: 10.1200/JCO-25-03120
-
O'Connor OA, Ma H, Chan JYS, Kim SJ, Yoon SE, Kim WS. Peripheral T-cell lymphoma: From biology to practice to the future. Cancer Treatment Reviews. 2024;129:102793. DOI: 10.1016/j.ctrv.2024.102793 • PubMed: 39002211
-
de Leval L, Nairismagi ML, Parrens M, et al. New insights into the biology of T-cell lymphomas. Blood. 2024. DOI: 10.1182/blood.2023021787 • PubMed: 39213420
-
Horwitz SM, Nirmal AJ, Rahman J, et al. Duvelisib plus romidepsin in relapsed/refractory T cell lymphomas: a phase 1b/2a trial. Nature Medicine. 2024;30(9):2517-2527. DOI: 10.1038/s41591-024-03076-6 • PMCID: PMC11862811
Disclaimer: The image accompanying this article is for illustrative purposes only and does not depict actual experimental results, data, or biological mechanisms.