Pilot Study Shows Promising Results With CPX-351 in Relapsed/Refractory ALL

In Partnership With:

Partner | Cancer Centers | <b>Moffitt</b>

Bijal Shah, MD, MS, discusses the results of a phase 2 pilot study investigating CPX-351 in adults with high-risk relapsed/refractory acute lymphoblastic leukemia, as well as next steps for research in this patient population.

CPX-351 (Vyxeos), a liposomal formulation of cytarabine and daunorubicin provided encouraging activity and overall response rates in adults with high-risk relapsed/refractory acute lymphoblastic leukemia (ALL) or mixed phenotype leukemia, according to data from a phase 2 trial (NCT03575325).1

The small open-label, single-arm pilot study featured 6 patients with B-cell ALL, including 1 with B-myeloid disease. Additionally, 5 patients had T-cell ALL, including 4 with an early T-cell phenotype. Moreover, a TP53 mutation was noted in 5 patients. Of the 10 evaluable patients, 3 had a complete response (CR) or CR with incomplete count recovery (CRi), and further research with this agent is warranted, according to Bijal Shah, MD, MS.

“We were happy with how CPX-351 performed in this pilot cohort,” Shah said. “This sets up a benchmark for expectations with single-agent CPX-351. It also tells us about who we’re going to be enrolling into relapsed and refractory trials as we go forward with the approvals of these novel agents in B-cell ALL.”

In an interview with OncLive®, Shah, associate member in the Department of Malignant Hematology at Moffitt Cancer Center, discussed the results of the phase 2 pilot study, as well as next steps in the patient population.

OncLive®: A presentation given by Marco Ruella, MD, of the University of Pennsylvania, at the 2021 ASH Annual Meeting and Exposition, discussed how gut microbes can affect responses to CAR T-cell therapy. How can this area be addressed in hematologic malignancies?

Shah: When we talk about how T cells kill, there’s the assumption that it’s largely a perforin/granzyme mediated cell-killing event. What was shared was that there may be other mechanisms involved here, including the induction of apoptosis in lymphoblasts, and what the investigators saw on a [CRISPR/Cas9] screen was that when you lose some of the proteins that help mediate apoptosis, you lose your CAR T-cell killing efficiency. There are 2 pieces to that.

One is that it seems like those cells can grow more rapidly. More proliferative leukemias are harder to kill. The other part of the equation is the T cells are less effective in eliminating those cells, despite perforin and granzyme production, and, classically, what we think of as T-cell mediated killing. Some of that may relate to the fact that the T cells are becoming exhausted as they proceed down this path and keep trying to kill this blast population that doesn’t want to accept defeat.

When we do genetic testing of leukemia, we’re not often looking at mutations or other changes along the apoptosis pathway. But it does suggest that we need to think more carefully, particularly in the more proliferative leukemias, to see whether there is an enrichment for those mutations that come with that enhanced resistance to CAR [T-cell therapy]. This fits into a whole category that we’re seeing as they relate to intrinsic CAR T-cell resistance.

We have talked about CD19 antigen downregulation, as well as CD58 and how that may affect CAR T-cell therapy and cancer interactions. We’re starting to now get a flavor for all these things, touching specifically on intrinsic features in lymphoma. I hope we’re also going to start to better understand tumor microenvironment features.

Dr Ruella talked about his studies looking at the microbiome and how that may affect T-cell biology. We know that antigen-presenting cells, or along that same spectrum, myeloid dendritic suppressor cell populations, can affect both the expansion and the contraction of T cells over time. There are a lot of moving parts here [in figuring out how we can] improve the activity of CAR T-cell therapies and the durability of cell killing.

[Regarding] toxicity, it seems like that may be more reaction to the T-cell growth rather than the T cells directly driving this phenotype. Targeting some of these pieces of the tumor microenvironment also improves toxicity, which, interestingly, may also improve T-cell activity and the durability of that activity.

It’s a rapidly evolving field, and it’s great to see. Hopefully we’ll get to a point where we can talk about CAR T-cell therapy as a replacement for chemotherapy. That’s the ultimate goal.

What was the rationale of the trial utilizing CPX-351 in patients with ALL?

Although I’m excited about [CAR T-cell therapy as a replacement for chemotherapy], it is true that there are other things that are happening, including a trial that we did examining [CPX-351].

This is the liposomal formulation of daunorubicin and cytarabine. We want to get rid of chemotherapy, but as it turns out, not everyone is an immunotherapy candidate. We’ve seen this in clinical practice. There are [patients] who are still on immunosuppressive therapy following an allogeneic stem cell transplant, and there are [patients] who don’t have B-cell ALL. T-cell ALL remains an area that we’re struggling to target from an immunotherapeutic standpoint. We wanted to try to understand what we could accomplish [with this agent].

This was a pilot study with 11 patients. The observations were interesting, and 1 of the key observations was who we enrolled. We have novel trials for ALL, including with CAR T-cell immunotherapy. Six [patients] who came onto [the study] had B-cell ALL, including 1 B-myeloid overlap.

Of the remaining 5 patients with T-cell ALL, 4 had an early T-cell phenotype. That tells us where some of our gaps are. When we look to the biology, 5 of the 10 evaluable patients had a TP53 mutation. [Additionally], 2 had [Philadelphia chromosome–like] changes, speaking to the high-risk nature of the patients who were enrolled to the study. They had a median of 3 prior lines of therapy, with a range of 1 to 7. Seven had primary refractory disease. When we talk about prior therapies for the B-cell ALL cohort, that included blinatumomab [Blincyto; n = 5], inotuzumab ozogamicin [Besponsa; n = 2], and CAR T-cell therapy [n = 1]. [Moreover], 3 patients had prior [allogeneic stem cell transplant]. This was a very heavily pretreated population, with [a] high [prevalence of] TP53 mutations and high rates of early T-cell precursor and B-myeloid phenotypes. [These were] the groups that you would expect since we don’t have great therapies for this group of patients.

What toxicities were seen in this trial?

We did see pancytopenia during induction as you would expect with cytarabine and daunorubicin. We did see febrile neutropenia in 9 [patients]. In terms of severe infections, it wasn’t what I would have predicted, particularly having been involved in the studies of CPX-351 in acute myeloid leukemia. We had 1 patient who [died] from pneumonia. This was a patient who had B-cell ALL, and she failed to collect for CAR T-cell therapy [since] she had low lymphocyte counts. We had hoped to intubate her and support her through the process, but she and her family decided to move to comfort measures in lieu of intubation.

A second patient developed grade 3 sepsis, which resolved without any significant issues. All other infections in the remaining patients were grade 1 to 2. The only atypical thing that we saw [in terms of] severe, unexpected adverse [effects] was 1 gastrointestinal bleed. This was in a patient who had prior allogeneic transplant, and there was a concern that this may have reflected some component of graft-versus-host disease, [GVHD], although we couldn’t confirm that. We had 3 grade 1 spontaneous subdural bleeds. [Those patients] all had low platelets, but we do see low platelets commonly in ALL without bleeding. This was a novel observation.

We had 1 patient develop veno-occlusive disease, but this is a patient who had prior inotuzumab [ozogamicin] and allogeneic transplant, so it was difficult to connect that with the CPX-351 therapy.

The infections that we saw were largely mild, meaning grade 1 to 2. Although we did see some unexpected cases of bleeding, [they] were incidentally appreciated in most cases and resolved.

The safety of this therapy was impressive in our patients with advanced ALL. The median time to neutrophil recovery was around 33 days. There were 2 patients who had refractory disease who didn’t recover their platelets, but among the remaining patients, [recovery] was around 30 days. We did allow consolidation, so just as in the ALL trial, patients received an induction dose and then subsequent consolidation doses where the CPX-351 was given at a lower dose.

We saw 1 episode of foot cellulitis. This was a patient who had recurrent foot cellulitis prior to enrolling on the study. [We also observed] 1 episode of myocarditis. That ultimately was not attributed to the study drug.

What were the results of this trial?

In terms of responses, 3 patients achieved a CR or CRi including 2 in the early T-cell precursor group and 1 who had B-cell ALL.

We did see responses in the TP53 cohort, so that was very encouraging. [Of those 4 patients], 1 had a CR/CRi, and 1 showed a reduction of blasts but not enough to meet the criteria for CR. One went on to receive consolidation, and one was bridged to a donor leukocyte infusion.

It seemed like we were succeeding in terms of supporting these patients. The median time to next therapy was about 2.5 months, so 76 days. The median overall survival for the patients was 218 days, so it seemed to come with some clinical benefit.

With regard to what happened when patients progressed, after CPX-351, we only saw responses to any subsequent therapy in 2 patients. The other 8 patients were refractory to every subsequent therapy that was administered, ranging from chemotherapy to targeted therapy. This speaks to the nature of the patients that were enrolled.

What are the next steps moving forward?

Next steps are to figure out what we can do with CPX-351. That may be combining it with novel agents like venetoclax [Venclexta] or even asparaginase [crisantaspase, erwinase]. It may also mean continuing to treat more intensively since we really didn’t see any significant toxicity in the consolidation phases.

Once patients moved to the lower intensity CPX-351, they were more likely to progress, if they didn’t go on to transplant. We would probably want to think about continuing the induction dosing for at least 1 additional cycle to help deepen responses for some of these patients.


  1. Shah BD, Rozario N, Turba EP, et al. A single-arm, open-label phase 2 pilot study of vyxeos (CPX-351) in adults with relapsed or refractory acute lymphoblastic leukemia. Blood. 2021;138(suppl 1):4399. doi:10.1182/blood-2021-151936