ASH: Bluebird's multiple myeloma CAR-T follow-up shows promise in phase 1

Bluebird bio has been trying something new with bb2121, its CAR-T therapy for multiple myeloma: processing it a little differently to boost the life span of the engineered T cells. Monday, it offered the first glimpse of how that approach is working in patients who have tried several other treatments without success. 

The phase 1 data, presented at the annual meeting of the American Society of Hematology, come from patients whose multiple myeloma has returned despite undergoing a median of six prior treatments, or whose disease did not respond to treatments. Of the patients on the lowest dose of bb21217, the treatment shrank tumors in 10 out of 12—83%—evaluable patients and eliminated them in four of them. Six of the 14 patients on the middle dose and four of the seven patients taking the highest dose saw their tumors shrink. 

The lowest dose, 150 million CAR-T cells per kilogram of body weight, kept cancer from growing or spreading for a median of 11 months, with one patient still going strong with a response lasting 20 months and counting. The data for the two higher doses, 300 million and 450 million CAR-T cells per kilogram, are still early, but none of the patients who saw their tumors shrink have had their disease get worse after a median follow-up of four months. 

“The 11.1-month duration of response is actually quite considerable given what’s currently out there,” bluebird bio Chief Scientific Officer Philip Gregory, D.Phil., told FierceBiotech. “If you think about single-agent approaches, if you think about the patients we’re treating, we would not expect most patients to be around within three to six months.” 

RELATED: ASH18: Bluebird’s anti-BCMA has a 'clear lead' in a crowded field 

The treatment is dubbed bb21217 because the manufacturing process cultures the CAR-T treatment bb2121, also called idecabtagene vicleucel or ide-cel, with a PI3K inhibitor called bb007. 

“When we create CAR-T cells, we do that by taking T cells out of the patient’s body and we do two things: We expand them in culture and during that expansion, we use a virus to integrate the CAR construct and express the molecule that will retarget the T cells to the tumor,” Gregory said. 

Culturing the CAR-T cells with the PI3K inhibitor during expansion, bluebird hopes, will limit the number of cells that differentiate into a more exhausted form of T cell—namely, T cells that don’t last very long in the body. Instead, the T cells that proliferate will be a more primitive, longer-lasting kind that have a “memory-like phenotype.” Importantly, the method doesn’t get in the way of expanding the cells. 

“We can have our cake and eat it: We can restrict differentiation without markedly changing the ability of the cells to expand,” Gregory said. 

The questions when bluebird designed the trial were, “Did we break it when we made this manufacturing change? Or do the T cells still do what they’re supposed to do—are they able to respond to the tumor and clear it in patients?” 

The safety and efficacy data from the study point to yes, bluebird and partner Bristol-Myers Squibb say. 

Of the 38 treated patients, 66% developed cytokine release syndrome (CRS), a dangerous side effect that happens when the CAR-T cells work too well and activate the immune system too strongly. Most of these cases were low-grade, requiring only moderate treatment or treatment to address symptoms like headache and fatigue. One patient needed aggressive treatment for CRS, and one patient died from the condition. The most common side effect was neutropenia, an abnormally low white blood cell count. 

“Everything we are seeing with ‘21217 is quite consistent with other CAR-T cell approaches across targets, including CD19 and BCMA,” Gregory said. “As is typical, a lot of the adverse events are associated with cytokine release syndrome. On one hand, that is a good thing: It means the T cells are getting to the tumor and beginning to kill the tumor and cytokines are being released in association with that. Of course, it can go too far on that scale and can get to very serious adverse events.” 

From a safety perspective, he said, the company hasn’t “broken” the T cells and they’re behaving similarly to other CAR-T treatments. From an efficacy perspective, it’s difficult to do a direct comparison to bb2121. But the company is finding a correlation between the number of longer-lasting memorylike T cells in the drug product and the ability of those CAR-T cells to expand in the patients, Gregory said. 

“Consistent with the hypothesis underlying the bb21217 program that memory-like phenotype T cells may survive longer in vivo, we have observed durable CAR-T cell persistence in patients with ongoing response at up to 18 months following treatment,” bluebird Chief Medical Officer David Davidson, M.D., said in a statement. “We are continuing to recruit additional patients in the study and performing ongoing assessments of the functional persistence of bb21217, as well as its potential correlation to durability of response.”

Whether boosting CAR-T cells’ life span equals longer responses and more effective treatments is a question the field has yet to answer. There are a few examples in the scientific literature, but time will tell whether bluebird’s approach with bb21217 will be an improvement over its predecessor, bb2121.

Jefferies analysts expect higher doses of the follow-up treatment to perform better than the lowest, 150 million dose.

"Collectively, the clinical profile of 150M cell dose cohort seems undifferentiated from that of bb2121. We await add'l updates from the 450M cell dose cohort to assess if central memory T-cell phenotype could be translated to improved clinical outcome," they wrote in a note on Tuesday.

Editor's note: This story was updated with comments from an analyst note.