Ryan Teague and Justine Kline recently put together a nice review of immune evasion in acute myeloid leukemia (AML). The open access paper is available online (http://www.ncbi.nlm.nih.gov/pubmed/24353898). These authors have particular interest in tumor escape from immune surveillance by two interesting mechanisms. The first is termed T cell exhaustion, and refers to a non-responsive state induced in CD8+ (cytotoxic) T cells. The second is immune suppression, mediated by TGFbeta and regulatory T cells (Tregs). Other means used by tumor cells to avoid the immune system include deactivation by co-opting signals that directly shut down immune responses, such as PD-1 and other signaling mechanisms.
Why the interest in immunotherapy for such an aggressive cancer? There are a number of good reasons. First I think it is fair to state that targeted therapeutics (small molecule drugs) and antibodies (mAbs, ADCs, bispecifics) have yet to achieve a breakthrough in AML. The best of these drugs, even in combination, are only modestly effective. The second reason, implicitly recognized by the T cell engaging bispecific antibodies (BiTEs, DARTs) and by the still nascent CAR-T cell engineering technology, is that there is evidence to suggest that AML can be controlled by an effective immune response. This evidence comes from the leukemia transplantation field. As Teague and Kline state:
“Treatment with modern chemotherapy regimens often induces complete remission, but a majority of patients will ultimately relapse … it has been recognized that allogeneic stem cell transplantation can be curative for some patients with AML … derived from the so-called graft-versus-leukemia effect thought to result … Unfortunately, only a minority of patients with AML are candidates for this procedure.”
Those who are familiar with allogeneic SCT will further recognize that this is a risky procedure that can outright fail. So, are there safer or more direct ways to harness an anti-tumor immune response?
Novel therapeutics developed to stimulate anti-tumor immunity include the CTLA4 antagonist mAb, ipilimumab (Vervoytm; Bristol Myers Squibb (BMS)), approved for use in refractory or non-resectable melanoma. BMS is also developing the anti-PD1 mAb nivolumab, and combination trials with ipilimumab are underway. Other anti-PD1 and anti-PDL1 antibodies in advanced development for a variety of tumor types include MK-3475, submitted last month for FDA approval for the treatment of advanced melanoma, MPDL3280A (Roche), MEDI4736 (Astra Zeneca), and others. These are critically important therapeutics in hematological cancer and solid tumors. The potential breadth of applications is illustrated by the announcement last week the Merck will seek collaborative partnerships to develop MK-3475 in combination therapies. Merck will partner with Pfizer to investigate combination therapy in a phase 2 renal cell carcinoma (RCC) trial with the VEGFR inhibitor axitinib (Inlytatm). Merck will also partner with Pfizer for a phase 1 trial(s) using MK-3475 with the agonist anti-41BB antibody PF-2566, in multiple cancers. Readers will note that 41BB signaling is a critical component of the CAR-T T cell engineering technology. The collaboration with Incyte is also a dual-immunotherapy approach, as MK-3475 will be combined with INCB24360, an IDO inhibitor, in a phase 1 non-small cell lung cancer (NSCLC) trial. IDO is secreted by tumor cells, is a mediator of T regulatory T cell activity, and in AML is associated with poor prognosis. With Amgen, MK-3475 will be used in combination with the oncolytic viral therapeutic T-VEC, which induces tumor cell death and stimulates anti-tumor immunity.
The point of all this is to illustrate that for difficult cancers – melanoma, RCC, NSCLC – its not going to be easy, and combinations of novel therapeutics will have to be utilized. AML is a very difficult cancer. With this in mind we can look at the state of immunotherapy drug development in AML.
The Teague and Klein review goes into considerable detail on this subject, so we’ll just hit a few highlights and then see if we can update the storyline. A point the review makes that I didn’t fully appreciate is that AML tumor cells (and many others) can downregulate MHC Class I and II, making the tumor cells difficult for the immune system to recognize in the context of allogeneic SCT. This fundamental type of immune evasion may be difficult to circumvent. Other mechanisms of immune evasion used by AML include expression of PD-1L on the tumor cells, which effectively shuts down tumor infiltrating T cells that express PD-1, the PD-L1 receptor and mediator of a potent signaling response that downregulates T cell activity. AML tumor cells also express B7 family proteins B7-1 and B7-2,that bind to CTLA4, another downregulatory receptor. Clinical trials enrolling AML patients for treatment with therapeutics such as ipilimumab, nivolumab etc are described in the review. Its sufficient to point out that the effort to use these therapeutics for AML is in its very earliest stages.
A few recent observations point to other immune evasion strategies that night be productively targeted in AML.
Several preclinical studies have identified co-expression of TIM-3 and PD-1 as markers of CD8+ T cell “exhaustion”, and have likewise identified PD-L1 and galectin-9 (a putative TIM-3 ligand) on AML patient cells. TIM-3 is yet another receptor on T cells that mediates downregulation of T cell activity. Other markers of AML cells from patients were recently described (https://ash.confex.com/ash/2013/webprogram/Paper56968.html).
Relevant proteins included B7-2 (CD86), B7-H3 (CD276) and PD-L1. Patients with very high expression of both B7-2 and PD-L1 had worse overall and relapse free survival. HVEM, a receptor for several critical immune proteins including LIGHT, CD160, and BTLA, was expressed on a subtype of AML with relatively good prognosis. The author’s conclude ” that the profile of immune checkpoint molecules … correlates with molecular disease characteristics in AML and may even possess prognostic information, especially for relapse … (and) as therapeutic targets with respect to boosting anti-leukemic immune responses.”
An example of such an approach is provided by Innate Pharma, which is developing an anti-KIR antibody, lirilumab. KIR negatively regulates NK cell anti-tumor activity. A phase 1 trial in AML is continuing (https://ash.confex.com/ash/2013/webprogram/Paper59174.html). Preclinical data support the use of this mAb in combination with the cytotoxic anti-CD20 mAb rituximab in lymphoma. One might envision a similar approach using a cytotoxic mAb targeting AML such as the anti-CD33 mAbs discussed in part 2. Another possibility are the anti-CD38 mAbs. Second generation CD38 mAbs with improved cytotoxic activity are under intensive development for multiple myeloma by Sanofi (mAb SAR650984), Jannsen (daratumumab aka HuMax CD38) and MorphoSys (mAb MOR03087).
Another example is CoStim Pharma, bought today by Novartis. In their portfolio are novel immunotherapeutic mAbs, including TIM-3 antagonist mAbs. Novartis is moving quickly here to beef up its immunotherapeutic pipeline, which it can now develop in parallel with the U Penn CAR-T technology. Another local, private immunotherapy company is Jounce Therapeutics.
As we have also seen in parts 1 and 2, drug development for AML lags significantly behind other leukemias, lymphomas, myelomas, and the like. However, targeted therapeutics such as the tyrosine kinase inhibitor sorafenib, HDAC inhibitors vorinostat and panobinostat, and proteosome inhibitors bortezomib and carfilzomib hold some promise. The FLT3 and c-Kit targeting agents seem less likely to provide meaningful long-term benefit, although we’ll see what the combo trials brings. While it is too early to assess the CAR-T technology, the bispecific modalities, or immunotherapies in AML, the cytotoxic mAbs and ADCs should have a prominent role in controlling this aggressive disease.
We asked in Part 1 who the winners would be in 5 years. Looking over the landscape of therapeutics its pretty clear that winning will require collaboration among companies. With that said those companies with the biggest concentration of effort in AML include Merck, Onyx, Novartis, Amgen and perhaps Seattle Genetics. Given their past successes we can be hopeful that several of these companies will succeed in establishing breakthrough treatments for AML. In the end, patients should benefit the most from all of this activity. Perhaps stockholders will also benefit. With this in mind we note that Onyx probably has the most to gain (or lose) in this indication.