Last week, Seattle Genetics (SGEN) announced that it had received notice from the U.S. FDA that a clinical hold or partial clinical hold has been placed on several early stage trials of vadastuximab talirine (SGN-CD33A) in acute myeloid leukemia (AML). The clinical holds were initiated to evaluate the potential risk of hepatotoxicity in patients who were treated with SGN-CD33A and received allogeneic stem cell transplant either before or after treatment. Six patients have been identified with hepatotoxicity, including several cases of veno-occlusive disease (VOD), with four fatal events.
The news has relevance to Actinium Pharmaceuticals (ATNM) because Actinium has two clinical stage candidates, Iomab-B and Actimab-A, for the treatment of AML. Actinium’s Iomab-B targets similar pre-transplant patients, although the Phase 3 Iomab-B program is open to sicker (less fit) subjects. Actimab-A, which consists of the humanized monoclonal antibody lintuzumab, is similar to Seattle Genetics’ vadastuximab in that both antibodies target CD33, a transmembrane receptor leukocyte antigen primarily expressed on cancer cells of myeloid lineage. Yet, there are important differences that investors need to be made aware of that lower the risk of hepatotoxicity and VOD with Actinium’s two programs. I discuss this issue below.
Seattle Genetics’ SGN-CD33A is an engineered antibody-drug conjugate (ADC) designed to deliver the cytotoxic agent pyrrolobenzodiazepine (PBD) dimer to myeloid leukemic cells. The antibody portion, vadastuximab, has engineered cysteines (EC-mAb) that link to the PBD dimer. Just like Actinium’s lintuzumab, vadastuximab binds to the CD33 protein on the myeloid leukemia cell, causing the complex to be internalized and trafficked to the lysosome. Once inside the cancer cell, PBD is released and binds to specific sites on the cellular DNA that ultimately lead to cell death.
Actimab-A consists of a similar monoclonal antibody to vadastuximab called lintuzumab, only Actinium uses the short-ranged (50-80 µm), high-energy (~100 keV/µm) alpha particle–emitting radioactive isotope Ac-225 to kill the cancer cells rather than the cytotoxic agent, pyrrolobenzodiazepine. Ac-225 is a far more effective cancer cell killing agent than PBD. PBD must be built up inside the cancer cell to sufficient concentration (picomolar) to interfere with DNA processing and elicit cell death. Conversely, individual Ac-225 ions can kill by proximity alone.
The result is that Actimab-A is far more powerful than SGN-CD33A, and thus can be dosed at lower concentrations. Actimab-A has also been shown effective without the need for pre-treatment with hypomethylating agents (HMAs) such as decitabine or azacitidine. HMA’s are typically used to increase the cell surface density of the CD33 transmembrane receptor. Both Seattle Genetics and Boehringer Ingelheim are deploying this strategy with their respective CD33 targeting mAbs. Use of these toxic agents, which has been associated with increased hepatotoxicity, is not necessary with Actimab-A.
SGN-CD33A Hepatoxicity – Not A Surprise
Seattle Genetics’s CD33A is not the fist CD33-targeting antibody-drug conjugate (ADC) to show increased risk of hepatotoxicity or fatal veno-occlusive disease (VOD). Pfizer’s Mylotarg® (gemtuzumab ozogamicin), approved for the treatment of AML in 2000, showed hepatotoxicity during clinical studies and received an FDA “Black Box” warning for increased risk of hepatotoxicity, including severe VOD, on the label. Mylotarg was a similar ADC to SGN-CD33A in that the CD33-targeting mAb gemtuzumab was linked to the cytotoxic agent calicheamicin. It seems that relapsed / refractory AML patients simply cannot tolerate these cytotoxic payloads on top of reduced intensity or non-myeloablative conditioning before allogeneic stem cell transplant. It’s too much assault on the liver. Mylotarg was eventually withdrawn from the market by Pfizer in 2010.
It is also not surprising to see SGN-CD33A run into trouble given the safety issues seen in previous clinical studies. As a reminder, Seattle Genetics is studying SGN-CD33A in a Phase 3 clinical trial combined with HMA’s called CASCADE. CASCADE is looking at the potential for CD33A to extend overall survival in elderly patients with newly diagnosed AML. This is similar to the indication for Actimab-A. The dose being studied in the Phase 3 trial is only 10 mcg/kg. The 10 mcg/kg dose was towards the bottom of the range tested in early-stage studies (5 mcg/kg to 60 mcg/kg), suggesting that doses above this level may be intolerable.
Seattle Genetics early-stage programs placed on full or partial hold by the U.S. FDA are looking at the pre-transplant population, similar to the indication for Actinium’s Phase 3 Iomab-B. Despite the fact that this program was only open to ECOG status 0 or 1 patients (i.e. healthier / more fit patients), the risk of hepatotoxicity, including severe VOD, was too great to continue.
Actimab-A & Iomab-B – No Significant Risk of Hepatotoxicity Seen To Date
I spoke briefly with Actinium after Seattle Genetic’s press release, and they confirmed to me that no severe hepatotoxicity or VOD has been seen to date with Actimab-A or Iomab-B. Actimab-A and its predecessor, Bismab-A, have been studied in roughly 90 patients with no significant issues of hepatotoxicity or VOD seen to date.
Iomab-B targets the common lymphocyte antigen CD45 and has no cytotoxic agents processed by the liver. The drug is currently in a Phase 3 clinical trial called SIERRA (see my overview of the Phase 3 trial). Iomab-B has been studied previously in almost 300 patients and, despite the fact that SIERRA is open to sicker (i.e. ECOG status 2) patients than the recently halted Seattle Genetics programs, safety data to date has been very encouraging. Two independent data safety analysis of SIERRA are scheduled to take place in 2017.
Final Thoughts
The risk of severe hepatotoxicity or VOD with CD33-targeting drugs such as SGN-CD33A and Mylotarg® seems to result from the increased assault on the liver by cumulative cytotoxic agents such as pyrrolobenzodiazepine or calicheamicin, along with the use of concomitant hypomethylating agents and reduced intensity or non-myeloablative conditioning. It does not seem to be in any way due to targeting CD33 itself or inherent in the pre-transplant patient population. In fact, Actinium’s HuM195 conjugated alpha particle program, which includes Actimab-A, has been tested in nearly 90 patients and no severe hepatotoxicity or VOD has emerged to date (see my analysis of the recent Actimab-A data at ASH).
Similarly, Actinium’s Iomab-B has been studied in nearly 300 patients without substantive risk of severe hepatotoxicity or VOD. Thus, the Seattle Genetic news to temporarily halt clinical work in some studies with CD33A is not a negative reflection on Actinium. It might, in fact, instead be an opportunity. Accordingly, I continue to believe that Actinium is well positioned for 2017