PLoS One. 2016 Sep 9;11(9):e0162472.

Targeting Attenuated Interferon-α to Myeloma Cells with a CD38 Antibody Induces Potent Tumor Regression with Reduced Off-target Activity

Sarah L. Pogue1*, Tetsuya Taura1, Mingying Bi1, Yong Yun1, Angela Sho1, Glen Mikesell1, Collette Behrens2, Maya Sokolovsky3, Hussein Hallak3, Moti Rosenstock3, Eric Sanchez4, Haiming Chen4, James Berenson4, Anthony Doyle2, Steffen Nock1, David S. Wilson1

1 Teva Pharmaceuticals, Global Branded Biologics Division, Redwood City, California, USA

2 Teva Pharmaceuticals, Global Branded Biologics Division, Sydney, Australia

3 Teva Pharmaceuticals, Global Branded Biologics Division, Netanya, Israel

4 The Institute for Myeloma and Bone Cancer Research, West Hollywood, California, USA

 

Abstract

Interferon-α (IFNα) has been prescribed to effectively treat multiple myeloma (MM) and other malignancies for decades. Its use has waned in recent years, however, due to significant toxicity and a narrow therapeutic index (TI). We sought to improve IFNα’s TI by, first, attaching it to an anti-CD38 antibody, thereby directly targeting it to MM cells, and, second, by introducing an attenuating mutation into the IFNα portion of the fusion protein rendering it relatively inactive on normal, CD38 negative cells. This anti-CD38-IFNα(attenuated) immunocytokine, or CD38-Attenukine™, exhibits 10,000-fold increased specificity for CD38 positive cells in vitro compared to native IFNα and, significantly, is ~6,000-fold less toxic to normal bone marrow cells in vitro than native IFNα. Moreover, the attenuating mutation significantly decreases IFNα biomarker activity in cynomolgus macaques indicating that this approach may yield a better safety profile in humans than native IFNα or a non-attenuated IFNα immunocytokine. In human xenograft MM tumor models, anti-CD38-IFNα(attenuated) exerts potent anti-tumor activity in mice, inducing complete tumor regression in most cases. Furthermore, anti-CD38-IFNα(attenuated) is more efficacious than standard MM treatments (lenalidomide, bortezomib, dexamethasone) and exhibits strong synergy with lenalidomide and with bortezomib in xenograft models. Our findings suggest that tumor-targeted attenuated cytokines such as IFNα can promote robust tumor killing while minimizing systemic toxicity.

PMID: 27611189; DOI:10.1371/journal.pone.0162472

 

Supplement

Multiple myeloma (MM) is a serious blood cancer diagnosed in approximately 30,000 people in the U.S. each year. It is the second most common blood malignancy in the U.S. next to non-Hodgkin’s lymphoma [REF 1,2].  Its cause is unknown and there is no cure. This disease affects plasma cells, the specialized white blood cells responsible for producing antibodies. In MM, malignant plasma cells generate excess antibody proteins that accumulate in the bone marrow and blood and eventually lead to organ and tissue damage. Early stage MM is often symptom free, but as the disease progresses patients experience debilitating weakness, fatigue, weight loss, kidney problems, weakened bones and bone pain.

Current treatments to prolong survival include stem cell transplantation, high-dose chemotherapy, use of various immune function related drugs and, more recently, direct tumor targeting antibodies. Nearly all treated MM patients eventually relapse and become resistant to therapy, with the five-year survival rate remaining around 47% [REF 1,3,4]. Thus, it is critical to develop new drugs and treatments for this blood cancer.

Interferon-α (IFNα) is an immune-modulating molecule that belongs to a class of cell-signaling proteins known as cytokines. Cytokines bind specific cell surface receptors leading to a cascade of intracellular signals that alter target cell activity. Most cytokines have multiple actions depending on the target cell they bind to and the abundance of their receptors on the cell surface. For example, IFNα can both activate immune cells to eradicate tumors but it also kills tumor cells directly by inducing cell cycle arrest or apoptosis.

IFNα has a well-established cytotoxic effect on many tumor cells and it significantly improves survival rates for MM patients. In fact, it has been prescribed for nearly three decades to treat MM and other cancers [REF 5,6]. Nevertheless, clinical use of IFNα for MM has waned in recent years due to its intolerable systemic side effects. In one study, maintenance therapy with IFNα was discontinued in up to 37% of MM patients because of treatment toxicity [REF 5]. The most commonly reported side effects of IFNα treatment include extreme nausea, flu-like symptoms, and blood cell insufficiencies leading to anemia and bleeding disorders; anxiety and depression are also often reported [REF 6-9].

The problem with IFNα use is that relatively high doses of the drug are required to generate effective anti-tumor responses. As such, there is only a small window between an effective dose and toxic dose of IFNα in patients (i.e., narrow therapeutic index, or TI). We proposed to broaden the TI of IFNα and to improve its potential clinical activity in MM by employing a combination of genetic engineering and molecular targeting strategies. Our ultimate goal is to minimize the off-target toxicities of IFNα while retaining its potent anti-tumor cell activity.

Others have attempted to enhance the TI of cytokines by attaching them to tumor-targeting antibodies, creating what are called “immunocytokines” (Figure 1). Specific targeting of cytokines to tumor cells with an antibody would increase the local concentration of cytokine at the tumor site, theoretically decreasing the overall amount of drug required for effective therapy. However, the potential for systemic toxicity persists with this strategy because the circulating cytokine, even if bound to a tumor-targeted antibody, can still freely bind to its natural receptor on normal cells and activate cellular activities in healthy tissues that lead to the systemic side effects.

 

To address this off-target toxicity and thus improve the TI of immunocytokines, we proposed mutating the cytokine portion of the molecule so that its affinity for receptor would be significantly lower than that of wildtype cytokine (Figure 1). We reasoned that a genetically altered immunocytokine with greatly reduced affinity to its receptor would have reduced “off target” activity in normal, target negative tissues and thus less systemic toxicity. Such an attenuated immunocytokine, dubbed “Attenukine™”, would remain effective on tumor cells due to increased local concentration facilitated by antibody targeting to tumor antigen.

Figure 2 illustrates how we applied this strategy to generate an IFNα Attenukine™ to potentially improve IFNα treatment of MM. IFNα was genetically fused to an antibody that specifically binds CD38, a cell surface protein that is highly expressed on nearly all MM tumor cells and minimally expressed on normal cells. We also introduced a single point mutation to the IFNα portion of the molecule to cripple its ability to bind its receptor (IFNAR).  The resulting antibody-attenuated IFNα fusion protein has an approximately 200,000-fold lower potency than free, unmutated IFNα.   We hypothesized that with this mutation, the IFNα Attenukine™ would generate little to no signaling on normal cells but would retain potent tumor cell killing on CD38-positive cells due to the antibody targeting those cells. 

 

 

We tested the vitro anti-tumor activity of our IFNα Attenukine™, identified as CD38-Attenukine™, on a CD38-positive MM cell line, ARP-1. The CD38-Attenukine™ retained potent anti-proliferative activity on these CD38 positive myeloma cells relative to native IFNa (Figure 3), whereas, activity on CD38 negative (non-tumor) control reporter cells was reduced by at least 200,000 fold. These in vitro findings suggest that, in patients, this IFNα Attenukine™ would specifically target IFNα to the tumor and not stimulate normal, CD38 negative tissues, potentially preventing systemic toxicities.

 

 

To ensure that our IFNα Attenukine™ retained strong anti-tumor activity in vivo, we tested the CD38-Attenukine™ in a MM mouse xenograft model. Human NCI-H929 MM cells were implanted in CB.17 SCID mice that were subsequently treated twice weekly for four weeks with the CD38-Attenukine™, vehicle, and other controls. We were surprised at the profound anti-tumor activity exhibited by the CD38-Attenukine™ resulting in durable tumor regression and significantly increased survival time in treated animals (Figure 4A). In fact, by day 22, all mice (10/10) were completely tumor free and remained so for the duration of the study. Furthermore, when we compared the efficacy of the CD38-Attenukine™ to three standard MM therapeutics (dexamethasone, Revlimid®, and Velcade®) in the same in vivo model, we found that the CD38-Attenukine™ generated the strongest anti-tumor response and longest survival time of all compounds tested (Figure 4B). To our knowledge, the remarkable, durable anti-tumor activity of CD38-Attenukine™ is unsurpassed by any current small or large molecule tested to date in similar MM animal models.

 

 

Finally, we examined the potential toxicity of the CD38-Attenukine™ treatment on healthy immune cells in vitro and in vivo. Our results demonstrate that the CD38-Attenukine™ has very little activity in suppressing colony formation of normal human bone marrow cells. In contrast, native IFNa significantly inhibited bone marrow cell colony formation at doses 6000x lower than the CD38-Attenukine™ (Figure 4C). In addition, separate preliminary hematological studies using healthy cynomologus monkeys also suggest that a non-targeted, attenuated IFNα-antibody fusion protein minimally stimulated IFNa-related cell responses within normal tissues in primates compared to wild type IFNa controls (data not shown).

Importance of this study: Our study describes a novel, MM tumor-targeted IFNα Attenukine™ that harnesses the potent anti-tumor activity of native IFNα while reducing the off-target activity responsible for the well-documented toxic side effects of IFNa. We demonstrate robust tumor inhibition both in vitro and in vivo. These results suggest that the CD38-Attenukine™ may provide a potentially broader therapeutic index (TI) and improved therapy for MM patients compared to native IFNa or to conventional, non-attenuated IFNa immunocytokines.  This is the first example of what we hope will be a generalized approach to enhancing the anti-tumor activities of therapeutic antibodies and improving the therapeutic indices of cytokine-based therapeutics.

 

References

  1. Institute NC. SEER stat fact sheets: myeloma [cited 2015 September 10]. Available from: http://seer.cancer.gov/statfacts/html/mulmy.html.
  2. Raab MS, Podar K, Breitkreutz I, Richardson PG, Anderson KC. Multiple myeloma. Lancet. 2009;374(9686):324-39. doi: 10.1016/S0140-6736(09)60221-X. PubMed PMID: 19541364.
  3. El-Amm J, Tabbara IA. Emerging Therapies in Multiple Myeloma. Am J Clin Oncol. 2013. doi: 10.1097/COC.0b013e3182a4676b. PubMed PMID: 23934133
  4. Siegel R, Ma J, Zou Z, Jemal A. Cancer statistics, 2014. CA Cancer J Clin. 2014;64(1):9-29. doi: 10.3322/caac.21208. PubMed PMID: 24399786
  5. Schaar CG, Kluin-Nelemans HC, Te Marvelde C, le Cessie S, Breed WP, Fibbe WE, et al. Interferon-alpha as maintenance therapy in patients with multiple myeloma. Ann Oncol. 2005;16(4):634-9. doi: 10.1093/annonc/mdi125. PubMed PMID: 15741226.
  6. Al-Zahrani H, Gupta V, Minden MD, Messner HA, Lipton JH. Vascular events associated with alpha interferon therapy. Leuk Lymphoma. 2003;44 (3):471-5. doi: 10.1080/1042819021000055066. PubMed PMID: 12688317
  7. Kirkwood J. Cancer immunotherapy: the interferon-alpha experience. Semin Oncol. 2002;29(3 Suppl 7):18-26. PubMed PMID: 12068384.
  8. Ludwig H, Cohen AM, Polliack A, Huber H, Nachbaur D, Senn HJ, et al. Interferon-alpha for induction and maintenance in multiple myeloma: results of two multicenter randomized trials and summary of other studies. Ann Oncol. 1995;6(5):467-76. PubMed PMID: 7669712.
  9. O’Brien S, Kantarjian H, Talpaz M. Practical guidelines for the management of chronic myelogenous leukemia with interferon alpha. Leuk Lymphoma. 1996;23(3-4):247-52. doi: 10.3109/10428199609054827. PubMed PMID: 9031105.

 

 

Multiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier SchönmannMultiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier Schönmann