Determining the Molecular Profile of BTK Inhibitor Responsive DLBCL Tumors

In an interview with Targeted Oncology, Mark Roschewski, MD, discussed precision medicine in diffuse large B-cell lymphoma and using acalabrutinib in a pre-chemotherapy window to better understand its potential benefit.

Responses among those treated with Bruton’s tyrosine kinase (BTK) inhibitors differ across subtypes of diffuse large B-cell lymphoma (DLBCL), according to Mark Roschewski, MD.

Although the combination of ibrutinib (Imbruvica) and rituximab (Rituxan), cyclophosphamide, doxorubicin, vincristine, prednisone (R-CHOP) does improve survival in patients, the benefit is limited by the toxicity of the regimen. Similarly, acalabrutinib (Calquence) has activity in DLBCL, according to research.

However, what remains unknown is how patients with DLBCL who have varying molecular profiles will respond to the therapy. Investigators sought to provide a better understanding of acalabrutinib treatment for DLBCL by administering the agent 14 days before frontline therapy.

In a recent study, a total of 34 patients were enrolled, and complete responses (CRs) to acalabrutinib were observed in 27. Two patients relapsed after the achievement of a CR. The estimated progression-free survival rate observed after a median follow-up of 9.2 months was 84.9% (95% CI, 58-95). Fifty percent and 35% of patients, respectively, experienced grade 3 or 4 neutropenia. Grade 3 thrombocytopenia occurred in 22%, and the events were grade 4 for 12% of patients. There were no increases in infections, atrial fibrillation, or bleeding in the patients treated with acalabrutinib in the study.

In an interview with Targeted Oncology™, Roschewski, senior clinician, Lymphoid Malignancies Branch, Center of Cancer Research, National Cancer Institute, discussed precision medicine in DLBCL and using acalabrutinib in a pre-chemotherapy window to better understand its potential benefit.

TARGETED ONCOLOGY™: Can you explain the differences between the DLBCL subtypes?

Roschewski: During ASH 2021, we talked a little bit about precision medicine and large B-cell lymphoma. What we know is that our preclinical models suggest that there are certain subtypes of large cell lymphoma that will preferentially respond to targeted agents. We've seen that in clinical studies, most of those clinical studies have been in the relapse setting. And so those are slightly different patient populations. In the untreated patient population. We do have randomized studies that don't show much difference with the use of targeted agents. But recently, our group did show that there were specific genetic subtypes of diffuse large B cell lymphoma that seem and in younger patients that seemed to have a benefit when a Bruton it was added to our job. So, we don't know yet enough about the individual response to targeted agents within subtypes of lymphoma.

What efficacy and safety findings have been reported so far with the combination of ibrutinib and R-CHOP in DLBCL?

In untreated large B-cell lymphoma, the most important study has been a randomized study known as the PHOENIX trial. The primary end point of that study was negative, meaning we could not see a benefit of ibrutinib with our R-CHOP across diffuse large B-cell lymphoma. But what was very interesting is that younger patients under 16 did seem to have a benefit, according to a subgroup analysis. More recently, our group published that there were 2 genetic subtypes of large B-cell lymphoma in that younger patient population, these are known as N1 and MCD that did seem to have a benefit. So, 1 of the problems we have is the heterogeneity of large cell lymphoma. But at this point, it's unclear who should get those medicines just because the data isn't sufficient.

What makes acalabrutinib a good treatment candidate for certain DLBCL subgroups?

The BTK inhibitors do seem like they have a role and they work very quickly. There are many patients that benefit from them, which we can see in the relapse setting. Acalabrutinib is very similar to ibrutinib, but it has more specificity. So, it binds to te similar epitope of the receptor. But it gets higher concentrations because it's given twice a day. So, the BTK occupancy is higher. Because of that selectivity, there is this theoretical benefit that it will have fewer off-target effects. So that will limit potentially some of the toxicity that is seen, and there is some evidence that that's true.

What were the key goals of your phase 2 study of acalabrutinib prior to frontline therapy in untreated patients with aggressive b-cell lymphoma?

It is important understand that we have a very unique study design. So, since we're interested in understanding, primarily, the role of the calibrated what we did was we implemented a window design. What happens is we first give the acalabrutinib by itself to all comers for up to 14 days. Because the medicine works very quickly, when patients benefit from it, we can see reductions in tumors very fast. The main focus in research is what are the types of patients that benefit from this acalabrutinib and we're going to be doing molecular profiling of these tumors, and we're going to be comparing tumors that respond to acalabrutinib of compared to those that don't. So, that's the primary goal.

Now, from a clinical perspective, what happens is, if we see a response, and we define that as only a 25% reduction, so that's less than we might think about in conventional response criteria. But if there's only a 20%, at least a 25% reduction, then we take that to mean the tumor is at least somewhat responsive to acalabrutinib, and then they get that acalabrutinib with their chemotherapy. But if there is no response in the window, then they get treated with chemotherapy alone. So, it also has a response adaptive design, and in that way, we can justify giving acalabrutinib to patients because we know they have a benefit, and we don't have to take the risk of the toxicity in patients that we don't see any benefit in the window.

Can you discuss the results you recently presented during the ASH Annual Meeting?

This was an interim analysis, and overall, we're going to enroll up to 100 patients. So far, we've enrolled 40. What we've seen so far is that in the 39 patients that have completed the window, 18 of them have responded to acalabrutinib, while 21 have not. So, it's about half and half. Then, we will look very closely at the genetic profile of these tumors both by cell of origin, as well as Han's criteria, and genetic subtypes. Now, we take all subtypes of large cell lymphoma on this study, except for primary mediastinal B-cell lymphoma. So, we took an agnostic approach. In other words, a preceding hypothesis would be that the patients with ABC large cell lymphoma or non-GCB lymphoma would be the ones that are most likely to respond. But that's not what we've observed. So far, what we've seen is that our responders have been enriched for patients with GCB. And this has been confirmed by cell of origin testing by RNA sequencing.

Then we've also looked closely at the genetic subtype. As I mentioned, 1 of our preceding hypotheses would be that patients with MCD or N1 genetic subtypes would respond. But none of our responders have had those genetic subtypes. So, we do see that the responses to acalabrutinib are broader than we may have thought before this study.

Another important aspect of this study is we're interested in developing and understanding the role of circulating tumor DNA [ctDNA]. So, this is an analyte that may help us predict responses much quicker than imaging. In this 14-day window, we're actually sampling ctDNA every 7 days to see if early changes in ctDNA, we can predict response that we see on CT scan. So far, on the first patients that we've looked at, that's been very true, there's been a very tight correlation with how quickly the ctDNA goes down and CT scan response. What's really interesting is that we saw that we could predict CT response after 14 days, as early as only 7 days by ctDNA. So, it does suggest that changes or decreases are actually a sign that the patient is responding to the tumor. It's a pretty good correlation between that and kind of overall tumor burden. So, those are the main findings. And of course, our patients have done very well. The safety has been good on this study. We've only had a few events, and our survival right now is over 90%. So, I think it looks in a randomized way and uncontrolled way that our patients are benefiting.

In your opinion, what would a randomized phase 3 study for this strategy look like?

I think we have a little bit more to learn. One of the focuses, of course, is to try to home in very tightly on which patients benefit from acalabrutinib. If we can identify a group of patients that do not benefit from acalabrutinib, we could design a study where we just don't enroll those patients, so we don't take the risk of acalabrutinib in those patients. So, both of those are key objectives. And so, I think what we're focused on here is who's the right patient population.

So, if we had the ideal patient population if we knew how to identify them, and then we would only enroll those patients and then do a randomized study, probably acalabrutinib or not with conventional chemotherapy that would be one way to go. Now the ctDNA adds in a whole other aspect. The question then becomes, can we do a different type of randomized study where we're looking at responses early on with ctDNA to be the reason to randomize from there? So that's another potential study. I think we're already discussing these types of options, but it's a little early to know for sure, just because our data aren't quite mature enough. But we're starting to get a signal for the right way to move forward.

How would you explain the importance of doing gene expression profiling for these patients?

It's important to recognize that gene expression profiling is distinguishing that these are truly different tumors. So, we have to have an understanding of the underlying molecular biology of these tumors. Now, does that currently help us in the clinic select which treatment to use? No, not yet. Part of the reason it doesn't is that the story is much more complex than just gene expression profiling. So, we've been interested in this question for a long time. And the most recent papers we've done show us that these genetic subtypes sort of cross over with the gene expression profiling subgroups. There are also some tight correlations, and there are some that aren't. So, I do think it's really important with therapy, particularly new therapies to know which groups of patients benefit the most, so they can be prioritized. We have all these targeted agents and immunotherapy agents. The disease is so heterogeneous, that if we don't have a nuanced understanding of which patients benefit, then we're going to be guessing in the clinic, or we're going to be relegated to only answering important questions with randomized studies. Now, randomized studies are important. But we also have to have a better understanding of how we approach individual patients. It takes a long time to get there. But I think every study that tests new agents should be also reporting how it responds within these subgroups of patients and genetic subtypes. I think that's the future.


Roschewski M, Phelan JD, Pittaluga S, et al. Phase 2 study of acalabrutinib window prior to frontline therapy in untreated aggressive b-cell lymphoma: preliminary results and correlatives of response to acalabrutinib. Presented 2021 American Society of Hematology Annual Meeting & Exposition; December 11-14, 2021; Atlanta, GA. Abstract 524.