Impact Of Elotuzumab Therapy On Circulating and Ex Vivo Activated/Expanded Autologous Natural Killer (Auto-ENK) Cell Activity

CS1 is an ideal target for multiple myeloma (MM) therapy as it is highly expressed on MM while having a very limited expression profile in normal tissues.  Elotuzumab (elo), a humanized monoclonal antibody (mAb) targeting CS1, has an acceptable safety profile and clinical activity in relapsed/refractory MM when combined with the immune modulator lenalidomide (len) and low dose dexamethasone (dex).  The primary mechanism of action for elo is NK cell-mediated antibody-dependent cellular cytotoxicity.  Here we report on a patient who was given elo/len via a single patient IND 6 months after receiving therapy with ex vivo activated auto-ENK cells and low dose IL2 as previously described (Szmania et al, Blood ASH Annual Meeting Abstracts 2012;120:1912).  The patient had relapsing GEP70 high-risk MM with cytogenetic abnormalities and had failed multiple lines of prior therapy including 3 auto-PBSC transplants and further salvage treatments including len, bortezomib, pomalidomide and carfilzomib.   Although ENK cell therapy did not induce a response, subsequent disease progression was slow.  IV elo was started 187 days after ENK cell infusion, and given every 14 days at the currently studied dose of 10mg/kg.  Len at 15mg/day was given on days 1-21 of a 28-day cycle.  Dex premedication (p.o. 38mg; IV, 10mg) was added after a grade 2 infusion reaction was observed to elo dose #1. While on the ENK cell protocol this patient had a dramatic increase in circulating NK cell counts peaking 9 days after infusion (6300 NK/µL, a 48-fold increase from baseline). Although still in the high range, NK cell levels at the time of elo treatment had normalized somewhat (539 NK/ml), and the cell surface expression of key activating receptors was consistent with a resting phenotype.  NK cell count remained stable after the first dose of elo (530 NK/ml) but subsequently dipped to 179 NK/µL after elo dose #2.   Since dex has been reported to affect NK cell counts, it is important to note that an additional dose was taken prior to elo dose #2 due to a travel delay (in total 66 mg of dex was taken on this occasion).  Circulating NK cells (collected pre-elo and 11, 25, and 57 days after elo dose #1) had similar low activity against auto-MM collected prior to elo treatment (effector:target ratio 10:1, 0-5% specific lysis) and killing against MM collected after 5 elo doses was only modestly increased (3-12%).  However, the same circulating NK cells exhibited significantly increased cytolytic ability when additional elo (10µg/mL) was added during the in vitro E:T co-incubation (3-11 fold increase in killing over isotype control, p=0.0008) suggesting that the MM targets were not saturated with mAb. Bound mAb may have been reduced in part during target cell isolation and freeze/thaw.  Freshly prepared auto-ENK cells exhibited an activated immunophenotype and induced significantly higher killing of pre-elo MM (45%) compared to non-expanded NK.  ENK killing was higher still against MM collected after 5 doses of elo (61%).  When elo was present during the assay, ENK demonstrated the most effective killing of auto-MM, reaching levels  equivalent to that of the NK sensitive target K562 (85% vs. 82% lysis).  Successful mAb therapy for MM is now moving forward as target antigens with selective, high and homogeneous expression, such as CS1, are identified.  However, the activity of responding effector cells is a critical issue to consider.  Inadequate NK cell count and activity level has been reported in MM and steroids typically given to debulk and preempt mAb-induced infusion reactions may exacerbate this problem.  Immunomodulatory agents given to enhance immune cell activity are not sufficient to reverse the negative effect of steroids.  We have previously shown that large doses of highly activated auto-ENK cells can be safely infused and that these cells expand further after infusion.  In this study we show that ENK cells have significant activity in vitro against auto-MM and that elo further enhances this activity.  Combination therapy incorporating saturating doses of mAb followed by infusion of NK effector cells with optimized activity against auto-MM is an innovative approach that warrants investigation.  Infusing highly activated effector cells after dex/elo may be one way to reap the benefits of combining these modalities while circumventing steroid-induced immune suppression.