Melanie Cole, MS (Host): [00:00:00] Welcome to the podcast series from the specialists at Penn Medicine. I'm Melanie Cole. And today, we're highlighting leukemia and the genomic determinants of treatment outcomes. Joining me is Dr. Matthew Connor. He's a medical oncologist and an Assistant Professor of Clinical Medicine at Penn Medicine. Dr. Connor, thank you so much for joining us today. I'd like you to start by telling us a little bit about the scope of the issue we're discussing here today. Tell us about the state of leukemia, the prevalence. What are you seeing in the trends?

Dr. Matthew Connor: Well, thanks so much for having me, Melanie. In the last couple of decades, we've seen very rapid advancements in our ability to sub-define types of leukemia based on molecular genetics. And thus, I think it's really important for us to recognize how this is both affecting [00:01:00] prognosis and treatment for our patients.

Host: Well then, let's talk about genomic determinants. How do you define them? And when we think of treatment outcomes in acute lymphoblastic leukemia or ALL, how does a specific genomic mutation determine or alter that treatment course?

Dr. Matthew Connor: Well, genomic testing is really important for risk stratification in adult ALL. Our current approaches include multimodal testing at time of ALL diagnosis, which is a combination of multi-panel gene fusion testing, next-generation sequencing, FISH, and conventional cytogenetics and PCR testing.

So, ALL can be defined first of all by lineage, with most patients having B-cell lineage acute lymphoblastic leukemia, and then a smaller proportion having T-cell lineage. After that, the most important genomic determinant, specifically [00:02:00] in B-ALL is the Philadelphia chromosome, a fusion protein between BCR and ABL1 present in about a quarter of adult patients with B-ALL. And this subset is treated in a very different way compared to other forms of ALL, incorporating oral therapies with tyrosine kinase inhibitors upfront and with continuing therapy.

Aside from that major genomic determinant in B-ALL, certain ALL genomic profiles are present that we refer to as poor risk, that are predictive of higher risk of relapse after upfront therapy. For instance, patients whose leukemia has fewer than the normal complement of chromosomes, referred to as hypodiploidy, or patients with rearrangements in certain oncogenes such as [00:03:00] KMT2A tend to have shorter periods of relapse-free survival with standard treatment.

Others like rearrangements between ETV6 and RUNX1 or hyperdiploidy, so a higher than normal complement of chromosomes, confer a more standard risk, which portends a more favorable treatment outcomes.

With ALL, the most important thing to determine a patient's treatment course is their initial response to upfront therapy, which we can test with ultrasensitive markers of measurable residual disease. But there's still a genomic component of risk stratification, and that does affect how we think about treatment for our patients. For instance, some with very high-risk mutations like KMT2As or low hypodiploidy may benefit from consolidative allogeneic transplant if they achieve remission, while patients with standard risk disease generally don't need such [00:04:00] therapy intensification unless they have evidence of relapse or refractory disease in the future.

And then, there's emerging evidence that suggests that some of our newer targeted agents in upfront therapy may portend better outcomes for patients with high-risk genomic subsets. But that may be a moot point because a lot of our more targeted therapies are actually moving into the frontline of therapy for ALL regardless of individual patient risk.

Host: Dr. Connor, speak about some of the most prominent actionable biomarkers that are currently being used to determine treatment, particularly in oncology itself.

Dr. Matthew Connor: As discussed before, some actionable biomarkers in ALL and AML simply tell us if a patient is high risk for relapse and drive our decisions about therapy intensification, for instance, with transplant. Increasingly though, actionable also now means targetable. As we understand more about phenotypic markers and genomic drivers of leukemia, new treatments have emerged that really help us tailor therapy based on our testing.

I alluded to tyrosine kinase inhibitors or TKIs earlier that we use for ALL with Philadelphia chromosome rearrangements to block proliferative signaling. This was really the first ever oncologic targeted therapy discovered decades ago. And we now have more advanced second and even third generation TKIs that we use for Philadelphia chromosome-rearranged ALL.

In AML, likely about 50 to 60% of adult patients will have an oncogenic [00:06:00] mutation in their AML for which we have targeted therapy options. So, these include mutations in FLT3, for which we offer TKI therapy both in the frontline and relapse setting; IDH, for which we have effective second-line therapies after relapse; and then, in patients with high-risk KMT2A rearrangements or patients with NPM1 mutations, we have oral therapy now with something called menin inhibitors, which, approved by the FDA last year as second-line treatment options, are actively being studied in the frontline, including here at Penn.

We're also seeing advancements in other oncologic subfields, with novel therapies for disease markers that are traditionally associated with poor prognoses, such as RAS pathway inhibitors in pancreatic cancer, and therapies targeting mutations in the tumor suppressor gene TP53, which is a huge unmet need in patients with leukemia. [00:07:00]

Host: So then, what do you anticipate will be the role of whole genome sequencing in the future as part of the therapeutic armamentarium for the blood cancers?

Dr. Matthew Connor: You know, the way that we evaluate the cancer genome in patients with blood cancer is strikingly more advanced than even 10 years ago. So here at Penn, it's standard to send targeted next-generation sequencing panels of hundreds of potential oncogenes at the time of diagnosis in any patient with blood cancer.

Now, while such advancements have really revolutionized how we assess disease risk and how we approach treatment, we also have to be cognizant of the amount of raw data that sequencing panels provide. For instance, many individuals have polymorphisms in genes on our sequencing panels that are likely inherited and may have no [00:08:00] clinical significance whatsoever. We see this very frequently.

Whole genome or exome sequencing is an even higher level of complexity.  Interpretation of those results has to be done with caution and has to be guided by clinical presentation. For instance, currently, such testing is mostly being used in patients who have suspicion of genetic cancer predisposition syndromes, so people who already have some kind of phenotypic clue as to what we might be looking for with whole genome or exome sequencing.

In the future, I think whole genome sequencing or exome sequencing is going to become more prominent. However, I think some obstacles are still present, which includes needing higher analytical sensitivity of these tests and standardization to make sure they're on par with our [00:09:00] current targeted panels. And then, of course, turnaround time, and cost would need to be comparable to what we have currently.

Currently, here, we can turn around targeted genetic sequencing within a couple of days, whereas exome sequencing or whole genome is still taking quite a bit longer to turn around. And it's important in our patients with leukemia to have early data to help us risk stratify and to help us determine if they're eligible for beneficial targeted treatments.

Host: Dr. Connor, switching gears here for a second to discuss mixed phenotype acute leukemia or MPAL. Tell us a little bit about what it is and who it affects.

Dr. Matthew Connor: Mixed phenotype acute leukemia is a very descriptive name for this entity, which is actually a difficult entity to diagnose. It's a rare type of acute leukemia that expresses phenotypic markers of both AML and ALL. So, this means that these express [00:10:00] both myeloid lineage markers like MPO, CD33 often, as well as either B-cell markers like CD19 or T-cell markers like CD3.

Less than five percent of patients newly diagnosed with acute leukemia are classified as MPAL. It's more common in men than women, and slightly more common in adults than children, but still quite rare. Unfortunately, it's a disease that is a very high-risk subtype of leukemia. Historically, the median survival for patients with mixed phenotype leukemias was less than a year. However, that's older data, and this is likely improving with modern treatment methods.

Host: Well then, what's the first-line approach to MPAL? Do you treat it more like AML or ALL? How do you treat a relapse if the disease initiated as ALL returns as AML or vice versa?

Dr. Matthew Connor: In true MPAL, [00:11:00] ALL-style induction therapy,  is generally recommended unless there’s some extenuating circumstance where that couldn't be offered. Studies, including a fairly recent large meta-analysis a few years ago, do suggest improved likelihood of achieving remission with an ALL-style therapy approach, which is generally multi-agent chemotherapy, corticosteroids, and probably with the addition of anti-CD19 therapy with blinatumomab, which is standard upfront therapy now for ALL.

The most important factor in determining the prognosis of MPAL is the ability of patients to undergo allogeneic transplant after achieving remission. And so, therapy that can lead to the highest likelihood of achieving a first complete remission is optimal here in order to get patients to allogeneic transplant.

When patients relapse with [00:12:00] MPAL or if they have refractory disease, then things can become a little bit more hairy because this is a disease that has a lot of plasticity. It can change its phenotypic expression. It can have varied genomic markers. So, it's very important to make sure that we're getting repeat assessments of disease with more tissue each time, anytime a patient has a relapse or suspicion for refractory disease, because they can have what's called lineage switch, where at a time of relapse or progressive disease, they can express a phenotype that's more consistent with myeloid leukemia or more consistent with a lymphoid leukemia. And so, we need to be fluid with our treatment decisions based on real-time data.

Host: Well, are there then significant genetic markers that you look for or mutations that might predict a better or worse outcome in MPAL?

Dr. Matthew Connor: The most important genetic marker in MPAL is similar to ALL, the presence or absence of Philadelphia chromosome. So, patients with Philadelphia chromosome rearranged disease are treated similarly to Philadelphia chromosome-positive ALL patients with incorporation of BCR-ABL targeting TKI therapy.

This was actually traditionally seen as a poor prognostic marker, having Philadelphia chromosome with MPAL. But with modern therapy approaches, survival in these patients actually is now comparable and perhaps superior to those who do not have the Philadelphia chromosome, specifically because of the incorporation of TKI therapy into treatment.

Host: Dr. Connor, how important are second opinions in MPAL, and where are patients best treated for the disease?

Dr. Matthew Connor: Well, given its rarity and the adverse prognosis, treatment at a specialty center is really essential for patients who have mixed phenotype leukemia. It's a particularly difficult diagnosis to nail down.

Seeking opinions at specialty centers that are familiar with this diagnosis, who have hematopathologists who are familiar with this diagnosis, is key. Most patients with MPAL should undergo allogeneic stem cell transplant if they can achieve a first remission. And that's a second reason why referral to a tertiary care center, specifically one that specializes in stem cell transplant early in their course, is key.

Additionally, clinical trials of targeted therapies such as for KMT2A with menin inhibitors have often included patients who have mixed phenotype acute leukemias. And so, second opinions at centers that can provide clinical trial access is also very important, since these diseases do have a high relapse rate.

Host: Thank you so much, Dr. Connor, for joining us today and sharing your incredible expertise for other providers. To refer your patient to Dr. Connor at Penn Medicine, please call our 24/7 provider-only line at 877-937-PENN, or you can submit your referral via our secure online referral form by visiting our website at pennmedicine.org/refer. That concludes this episode from the specialists at Penn Medicine. I'm Melanie Cole.