Chronic lymphocytic leukemia (CLL) and mantle cell lymphoma (MCL) are tumors of mature B cells and are closely related biologically and in clinical behavior. Both are currently incurable with chemotherapy. Median survival for patients with CLL is 10 years and for MCL patients ranges between 5 and 6 years. BCR signaling has emerged as the pivotal pathway in the pathogenesis of CLL and MCL. A major contribution from my group has been the first demonstration of active BCR signaling in CLL patients in vivo. Furthermore, we showed that BCR signaling and the consequent activation of the NF-B pathway occurs primarily in the lymph node microenvironment rather than in the peripheral blood or bone marrow. Small molecule inhibitors of BCR signaling have transformed clinical approaches. The most promising clinical results have been achieved with ibrutinib. Between 2012 and 2014, our phase 2 clinical trial with ibrutinib enrolled 84 patients. At the end of 2018, 41 (49%) patients remained on study. Twenty-one publications from my group report data from this clinical study or from laboratory investigations into specimens obtained from participating patients. With the median follow-up now exceeding 5 years, we were able to address long-term tolerability, duration of response, mechanisms of resistance, and impact on immune function. At the time of data cut for our updated report (December 31, 2017), median follow-up was 57 months. Four (4.8%) patients died on study due to reasons other than progressive disease; 3 from infections, and one due to sudden unexplained death. Twenty (23.8%) patients discontinued ibrutinib due to disease progression, and 5 (5.9%) due to AEs. The cumulative complete response (CR) rate was 9.9% at 2 years, and 28.4% at 5 years. Seven patients in CR achieved minimal residual disease (MRD) negativity in either blood or BM. The estimated 5-year PFS was 58.2% for the TP53 cohort and 81.2% for the elderly cohort (P=.026); and the median overall survival (OS) was 75.7% and 83.8% respectively. In both cohorts, PFS was superior for previously-untreated patients (Figure 4). In the TP53 cohort, the estimated 5-year PFS was 74.4% for patients with treatment-nave CLL compared to 19.4% for those with relapsed or refractory CLL (p=.0002), and OS was 85.3% versus 53.7%, respectively (p=.023). Results with ibrutinib monotherapy in these high-risk patients were clearly superior to what has been reported for chemoimmunotherapy and our study provides the largest experience of ibrutinib for first-line therapy of CLL with TP53 aberrations and can serve as a benchmark for alternative treatment approaches. To improve tolerability and minimize off-target effects of ibrutinib, ACP-196 (acalabrutinib), a more BTK selective inhibitor was developed. Taking advantage of our PDX model, we contributed pre-clinical data supporting the translation of acalabrutinib into clinical trials. Between 2015 and 2018, we enrolled 46 CLL patients into a phase 2 single-center, open-label, clinical trial of acalabrutinib. Correlative studies focused on detailed profiling of pharmaco-dynamic readouts in relation to the degree of BTK inhibition. Forty-six patients were enrolled and randomized to receive oral acalabrutinib 100 mg twice daily (BID, n=22) or 200 mg daily (QD; n=24). The ORR for patients who completed 6 months was 90% and not different between BID and QD groups. The study incorporated repeat blood (days 0, 3, 4, 5, 28) and tissue (days 0, and days 3, or 4, or 5;
aimi ng for 2 biopsies of either LN or BM for each patient) sampling to investigate 3 pharmacodynamic questions: a) does BTK target occupancy differ for cells in different anatomic compartments; b) how quickly is BTK resynthesized; and c) at what threshold of BTK occupancy is downstream signaling restored? To quantify the risk of disease progression on BTKis, we developed a clinical scoring system. Our NIH trial of ibrutinib monotherapy served as the discovery cohort and the model was then validated in the Pharmacyclics database of 607 CLL patients treated with ibrutinib on company-sponsored studies. In the discovery cohort, factors significantly associated with inferior PFS on multivariate analysis were; TP53 aberration, prior treatment, and high B2M. Clinical prognostic groups were based on the number of factors present at baseline: none or one factor (low risk), 2 factors (intermediate risk), all 3 factors (high risk). PFS and OS of the risk groups were significantly different for both models (p<.0001). The 3-year PFS was 51% for patients in the high-risk group and 85% for the low-risk group (Figure 6). The cumulative incidence of BTK/PLCG2 mutations in high and low risk groups was 56% and 19%, respectively. This simple prognostic model, based on 3 readily-available baseline factors, outperformed the CLL-IPI and can serve to stratify patients for risk-based treatment approaches. To assess immunocompetence in patients treated with BTK inhibitors, we conducted a pilot study using influenza vaccine and could show that patients on ibrutinib respond at least as well as untreated CLL patients to the vaccine, demonstrating that patients on ibrutinib can still mount a meaningful antibody response. Given that infections are a leading cause of death, this data have important preventive health implications. We more recently opened a new study using Shingrix, a varicella-zoster vaccine, and Heplisav, a Hepatitis B vaccine to further investigate immune function in CLL patients and will compare treatment-nave patients to patients being treated with kinase inhibitors. Disease progression in CLL patients treated with ibrutinib has been attributed to histologic transformation and acquired mutations in BTK and PLCG2. We explored clinical and genetic characteristics of CLL patients who progressed during treatment with single-agent ibrutinib on our phase 2 trial. Fifteen (17.9%) of 84 patients progressed at a median follow-up of 34 months (Figure 9). Histologic transformation occurred in 5 (6.0%) patients, all in the first 15 months on ibrutinib. In contrast, progression due to CLL was diagnosed at a median 38 months on study. At progression, mutations in BTK (Cys481) and/or PLCG2 were found in 9 patients. High sensitivity testing of stored samples identified the mutations up to 15 months before clinical progression (range 2.9-15.4 months). In 5 patients multiple subclones carrying different mutations arose independently, documenting subclonal heterogeneity in resistant disease. To define the evolutionary dynamics induced by ibrutinib at high resolution, we performed serial exome and transcriptome sequencing for 61 ibrutinib-treated CLL patients. The frequent serial sequencing of early on-treatment (prior to relapse) samples enabled us to directly compare the relative diminution of different clonal subsets in response to ibrutinib. Thirty (49%) of 61 CLL patients showed significant pre-relapse changes in CCF over time. Of the 30 CLL patients with significant early clonal shifts, a likely branched pattern was observed in 20, in which distinct clones appeared to reciprocally rise and fall, suggestive of shifts in dominance between the two sibling clones, while the remainder of evolving CLL patients demonstrated a linear evolution pattern. In conclusion, early clonal dynamics may reflect a greater evolutionary capacity, irrespective of specific driver alterations, setting the stage for the emergence of drug-resistant clones.
Showing the most recent 10 out of 74 publications
