Envisioning a Path Forward for PI3K inhibitors in Hematologic Malignancies

Basem Goueli, MD, PhD, MBA

PI3 kinase (PI3K) inhibitors are offering the option of objective responses and progression-free survival improvement for those who treat patients with hematologic malignancies. But, recently, the toxicity of the drugs class and limited overall survival benefit has made these agents less attractive treatment choices.

An increasing imbalance in OS data recently led the biologics license application and supplemental new drug application for the combination of ublituximab (TG-1101) and umbralisib (Ukoniq) seeking FDA approval for the treatment of chronic lymphocytic leukemia (CLL) and small lymphocytic leukemia (SLL) to be voluntarily withdrawn by the developer. Reportedly, missing or outdated survival data occurred for approximately 15% of patients.

The FDA’s Oncologic Drugs Advisory Committee (ODAC) also brought the topic of PI3K inhibitors into question during the April 21 ODAC meeting. After being asked to discuss observed toxicities in the drug class, research showing a detriment in OS, a narrow range between effective, and toxic doses, the members unanimously voted that randomized data should be required to support future PI3K-inhibitor approvals.

In an interview with Targeted Oncology™, Basem Goueli, MD, PhD, MBA, chief executive officer and founder of CancerLight and Cancer Clarity, a practicing hematologist/medical oncologist, and medical director at Xbiotech, discussed the recent crackdown on PI3K inhibitors and how this drug class may fit into future treatment for hematologic malignancies as well as solid tumors.

TARGETED ONCOLOGY: What are your thoughts on the recent withdrawal of the applications for ublituximab with umbralisib for CLL/SLL?

Goueli: This was merely a matter of time. The FDA has clearly put its foot down on the fact that PI3 kinase inhibitors are fraught with risk.

It's actually really sad that this happened because a lot of well-intentioned companies got burned by this. When idelalisib [Zydelig] came out about a decade ago everyone was very excited and the developer looked like they had a home run. Unfortunately, patients developed gastrointestinal perforations and it was horrifying. In 2012, I had an investigator-initiated study using lenalidomide [Revlimid] and idelalisib in markedly refractory DLBCL that Gilead was interested in funding. I treated 2 patients with the combination and the results were nothing short of miraculous. However, both patients had marked side effects which precluded me from pursuing it further.

What have you seen with PI3K inhibitors in your patients with hematologic malignancies?

Looking across this class of drugs, including alpelisib for PIK3CA-mutated breast cancer, copanlisib [Aliqopa] for follicular lymphoma, duvelisib [Copiktra] for CLL, idelalisib for CLL, and umbralisib for indolent lymphoma, we find these drugs are difficult to tolerate, albeit to different degrees. Accordingly, zandelisib recently failed to get fast-track approval designation as the FDA is now keenly aware of potential issues with PI3K inhibitors.

Thinking of PI3K inhibitors in solid tumors, are you seeing similar issues?

The only PI3K inhibitor, as you know, that is currently FDA approved in solid tumors is alpelisib [Piqray] for PIK3CA mutated, metastatic hormone-receptor positive, HER2-negative breast cancer. Approximately 40% of metastatic breast cancer patients have an actionable PIK3A mutation. To this end, for these patients alpelisib is given with fulvestrant and is a very common second-line treatment after patients fail a CDK 4/6 inhibitor and an aromatase inhibitor. Alpelisib was approved in this indication based on the SOLAR-1 study that showed a median PFS of 11 months for alpelisib plus fulvestrant relative to 5.7 months with fulvestrant alone. Median overall survival was significantly prolonged in the alpelisib/fulvestrant arm. Importantly, the benefit of alpelisib was restricted to patients that had a PIK3CA mutation, further demonstrating the critical importance of next generation sequencing in stage IV breast cancer.

The patients I've given alpelisib to have had mild-moderate difficulty tolerating the drug, but were able to stay on treatment.Indeed, there is no question drug specificity is important as alpelisib is a PIK3C alpha inhibitor, idelalisib is a delta inhibitor, copanlisib is an alpha and delta inhibitor, duvelisib is a delta and gamma inhibitor, and umbralisib is a delta inhibitor.Accordingly, it's not surprising these drugs have different toxicity profiles.

ODAC voted that future approvals of PI3K inhibitors must be supported by randomized data. What are your thoughts on their decision?

The brave men and women of the FDA are making decisions pertaining to drug approvals and denials are doing good work.It's very easy for people to criticize and second guess their decisions. They really are in a thankless profession but are doing an incredible job in my opinion.

Ultimately, I sincerely feel this decision was long overdue. There was no reason for umbralisib to be fast tracked for indolent lymphomas as copanlisib was already approved in relapsed/refractory follicular lymphoma, and it's difficult to surmise umbralisib offers a profound benefit that copanlisib doesn’t, despite the different specificities. However, we're splitting hairs here. I unequivocally applaud ODAC for having the courage to mandate randomized control studies be performed for all PI3 kinase inhibitors seeking approval in the future. I don't feel they had a choice with the idelalisib and umbralisib issues, but I unequivocally commend them regardless.

What do you think this means for future of the 3 PI3K inhibitors approved based on single-arm trials?

I think the days of PI3K inhibitors being approved based on single-arm phase 2 trials are essentially gone. However, it's important to note that alpelisib was just approved this month for PIK3CA-related overgrowth spectrum ­based on a single-arm clinical study of 37 patients, largely because the disease is so rare. Of course, I feel this was completely appropriate, but I do feel PI3K inhibitor single-arm trials will be generally insufficient to get them approved in the future.

It's worth nothing that there are numerous trials involving copanlisib, alpelisib, and duvelisib ongoing. Many of these trials are in solid tumors other than breast cancer, and involve combination therapy. I suspect they will almost all require validation with randomized control trials before culminating in an approval. However, there are some potential exceptions, including copanlisib in primary central nervous system lymphoma where it is being tested in combination with ibrutinib [Imbruvica], and we need any therapy we can get.

What do to see for the future of this drug class in hematologic malignancies?

It's disheartening this class of drugs has taken the beating it has. It's difficult to envision a path forward for PI3K inhibitors as single agent therapy in hematologic malignancies that are saturated with amazing drugs that are also well-tolerated, such as CLL and indolent B-cell lymphomas. They will continue to have a place in the markedly treatment-refractory patient who has no additional therapeutic options, but will likely not be used as single agent therapy ahead of Bruton’s tyrosine kinase inhibitors, BCL-2 inhibitors, CD20-CD3 bispecific T-cell engagers, that I feel will be approved, etc. However, the companies making these drugs are astutely aware of this and are trying to establish a foothold in hematologic malignancies through unique combination therapies.

For example, as I was recently asked to do an assessment of copanlisib ongoing trials, I can say the drug is being tested in first-line DLBCL with R-CHOP, marginal zone lymphoma with rituximab, refractory DLBCL with bendamustine/rituximab, Burkitt's and high-grade DLBCL with the DA-REPOCH regimen, untreated follicular lymphoma with rituximab, primary central nervous system lymphoma with ibrutinib, relapsed/refractory CLL with ibrutinib, relapsed/refractory DLBCL with venetoclax, and Richter's transformation with nivolumab.

Ultimately, like everyone in the field, I'm looking forward to the data from PIK3 inhibitors in all sorts of trial iterations, including the ones I just discussed.