Rate of discontinuation is low overall, with relapse-related discontinuations less frequent than nonrelapse related discontinuations. Richter’s transformation (RT) tends to occur earlier than progression with typical chronic lymphocytic leukemia (CLL).
eMethods. Ion Torrent Deep Sequencing
eTable 1. Studies Included in Analysis
eTable 2. Individual Variables Related to Disease Progression, Adjusted for Monotherapy
eFigure 1. Survival Following Ibrutinib Discontinuation for Reasons Other Than Relapse
eFigure 2. Survival Following Disease Progression on Ibrutinib
Customize your JAMA Network experience by selecting one or more topics from the list below.
Maddocks KJ, Ruppert AS, Lozanski G, et al. Etiology of Ibrutinib Therapy Discontinuation and Outcomes in Patients With Chronic Lymphocytic Leukemia. JAMA Oncol. 2015;1(1):80–87. doi:10.1001/jamaoncol.2014.218
The Bruton tyrosine kinase (BTK) inhibitor ibrutinib is effective in patients with chronic lymphocytic leukemia (CLL). Reasons for discontinuing therapy with this drug and outcomes following discontinuation have not been evaluated outside of clinical trials with relatively short follow-up.
To determine features associated with discontinuation of ibrutinib therapy and outcomes.
Design, Setting, and Participants
A total of 308 patients participating in 4 sequential trials of ibrutinib at The Ohio State University Comprehensive Cancer Center were included. These clinical trials accrued patients included in this analysis from May 2010 until April 2014, and data were locked in June 2014.
Main Outcomes and Measures
Patients were evaluated for time to therapy discontinuation, reasons for discontinuation, and survival following discontinuation. For patients who discontinued therapy because of disease progression, targeted deep sequencing was performed in samples at baseline and time of relapse.
With a median follow-up of 20 months, 232 patients remained on therapy, 31 had discontinued because of disease progression, and 45 had discontinued for other reasons. Disease progression includes Richter’s transformation (RT) or progressive CLL. Richter’s transformation appeared to occur early and CLL progressions later (cumulative incidence at 12 months, 4.5% [95% CI, 2.0%-7.0%] and 0.3% [95% CI, 0%-1.0%], respectively). Median survival following RT was 3.5 months (95% CI, 0.3-6.0 months) and 17.6 months (95% CI, 4.7 months–“not reached”) following CLL progression. Sequencing on peripheral blood from 8 patients with RT revealed 2 with mutations in BTK, and a lymph node sample showed no mutations in BTK or PLCG2. Deep sequencing on 11 patients with CLL progression revealed BTK or PLCG2 mutations in all. These mutations were not identified before treatment in any patient.
Conclusions and Relevance
This single-institution experience with ibrutinib confirms it to be an effective therapy and identifies, for the first time, baseline factors associated with ibrutinib therapy discontinuation. Outcomes data show poor prognosis after discontinuation, especially for those patients with RT. Finally, sequencing data confirm initial reports associating mutations in BTK and PLCG2 with progression and clearly show that CLL progressions are associated with these mutations, while RT is likely not.
clinicaltrials.gov Identifiers:NCT01105247, NCT01217749, NCT01589302, and NCT01578707
Chronic lymphocytic leukemia (CLL) is the most prevalent leukemia in adults and is not considered curable outside of allogeneic stem cell transplantation. Significant advances have been made in the therapy, notably the emergence of kinase inhibitors for patients with relapsed disease. Prior to US Food and Drug Administration (FDA) approval of ibrutinib and idelalisib with rituximab, standard therapy for relapsed CLL could be expected to induce response rates of 30% to 50%1-3 and progression-free survival (PFS) of generally less than 1 year. The phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3 kinase) delta isoform inhibitor idelalisib4,5 and Bruton tyrosine kinase (BTK) inhibitor ibrutinib have significantly improved the outcomes for these patients.
BTK is a critical kinase in the B-cell receptor signaling pathway. This pathway is amplified in CLL and results in amplification of proliferation and antiapoptotic signals.6-9 By inhibiting BTK, ibrutinib eliminates the activation of these prosurvival pathways8 and microenvironment survival signals.8,10 In patients, this translates into high clinical response rates and durable remissions. The phase 1 trial of ibrutinib in relapsed B-cell malignant neoplasms showed activity in a variety of diseases,11 and the benefit in CLL has been confirmed in phase 1b/212,13 and phase 3 trials14 of ibrutinib as a single agent. In the phase 3 study (RESONATE trial), ibrutinib was compared with ofatumumab. With a median follow-up of 9.4 months, the overall response rate (ORR) was superior with ibrutinib, but more importantly both progression-free survival (median, 8.1 months for ofatumumab vs “not reached” for ibrutinib) and overall survival (OS) (12-month estimates: 81% for ofatumumab and 90% for ibrutinib) were significantly improved with ibrutinib.14 The OS benefit is particularly impressive considering that follow-up was short, and 14 months following the beginning of enrollment, patients whose disease progressed with ofatumumab therapy were allowed to cross over to ibrutinib therapy.
Ibrutinib has also been studied in combination with chemotherapy and immunotherapy. While laboratory data suggested that combination with CD20 antibodies may be less effective because of interleukin 2–inducible T-cell kinase (ITK) inhibition,15 which impairs the natural killer cell antibody-dependent cellular cytotoxicity,16 trials of ibrutinib in combination with rituximab17 and ofatumumab18 have shown similarly impressive efficacy. It is yet to be determined whether a combination of ibrutinib with another active agent is superior to ibrutinib as a single agent. While the response rate to ibrutinib has been high, with therapy well-tolerated overall, some patients have experienced relapse, while others have been taken off therapy because of toxic effects or other reasons. Our group and others have shown that relapse in CLL can be mediated by at least 2 separate mechanisms.19 One is by mutations in BTK, which both decrease ibrutinib’s affinity for BTK and change the binding from irreversible to reversible. The other is through a variety of mutations in PLCG2, the immediate downstream target of BTK. We have identified a number of mutations in PLCG2 that potentially have gain-of-function effects and likely serve to allow the cell an ability to signal in the presence of ibrutinib. Herein, we describe the characteristics of patients who discontinue ibrutinib therapy for relapse and nonrelapse reasons, identify factors that may contribute to discontinuation, perform deep sequencing for known resistance mutations, and describe outcomes of patients after ibrutinib therapy discontinuation.
Risk factors for discontinuation of ibrutinib therapy due to reasons other than progressive disease include older age and greater number of prior therapies.
Risk factors for disease progression include BCL6 abnormalities on pretreatment fluorescence in situ hybridization and complex karyotype.
Disease progression on ibrutinib therapy includes Richter’s transformation (RT), which tends to occur early in treatment, and chronic lymphocytic leukemia (CLL) progression, which tends to occur later.
Patients who progress with RT have a median survival of 3.5 months after progression.
Acquired mutations in BTK and PLCG2 account for most CLL progressions during ibrutinib therapy.
All patients at The Ohio State University (OSU) Comprehensive Cancer Center who provided written informed consent and were enrolled in 4 sequential trials of ibrutinib in patients with CLL, approved by the institutional review board of OSU, were included in this analysis. Study designs are outlined in the eTable 1 in the Supplement. OSU 10032 (Pharmacyclics Inc [PCYC] 1102; NCT01105247) was a multi-institutional phase 1b/2 study of ibrutinib as a single agent in patients with relapsed/refractory or treatment-naïve CLL.12 OSU 10053 (PCYC 1109; NCT01217749) was a single-institution phase 1b/2 study of ibrutinib in combination with ofatumumab in patients with relapsed/refractory CLL.18 OSU 11133 (NCT01589302) was a single-institution phase 2 study of ibrutinib as a single agent in patients with relapsed/refractory CLL.13 OSU 12024 (NCT01578707 [RESONATE; a trademarked trial name by PCYC]) was a multi-institutional phase 3 study of ibrutinib as a single agent vs ofatumumab in patients with relapsed/refractory CLL.14 From this study, all patients initially assigned to ibrutinib therapy or who crossed over to ibrutinib therapy following progression of disease with ofatumumab were included.
DNA was extracted from cryopreserved cells using QIAmp DNA Mini kit (Qiagen). BTK (OMIM 300300) and PLCG2 (OMIM 600220) genes were analyzed using the Ion Torrent platform from Life Technologies. Details regarding library generation and sequencing can be found in the eMethods in the Supplement.
Time to discontinuation of treatment was measured from the first date of treatment with ibrutinib until the off-study date, censoring patients who had not discontinued ibrutinib therapy at the date of last contact; patients who went off study for transplant or continued treatment elsewhere (n = 7) were censored at that time. Median follow-up was calculated among all patients (n = 232) censored for time to discontinuation of treatment. Fine and Gray models of cumulative incidence20,21 were fit to identify variables associated with a particular failure type and in the presence of competing risks. Multivariable models of cumulative incidence were fit using forward selection, with variables no longer entering the model if P > .05, unless deemed clinically important. All models were adjusted for monotherapy with ibrutinib vs combination therapy with ibrutinib and ofatumumab. Other variables considered included age; sex; number of prior therapies; baseline lactate dehydrogenase (LDH) level; fluorescence in situ hybridization (FISH) abnormalities del(17p), del(11q), trisomy 12, del(13q); MYC abnormalities; BCL6 abnormalities; and complex karyotype and IGHV mutational status. Survival following discontinuation was calculated from the off-study date until the date of death from any cause, censoring patients at last contact. Survival estimates were calculated by the Kaplan-Meier method, and differences between curves were tested with the log-rank test. All tests were 2-sided, and statistical significance was declared at α=.05. The cmprsk package within TIBCO Spotfire S+ Version 8.2 was used to analyze time-to-event data in the presence of competing risks, whereas SAS statistical software Version 9.3 (SAS Institute Inc) was used for all other data analyses.
A total of 308 patients were included in this analysis, with data locked as of June 9, 2014. This included 237 patients treated with ibrutinib as a single agent in 3 clinical trials and 71 patients treated with ibrutinib in combination with ofatumumab. Clinical characteristics for all patients are given in Table 1. Three patients treated with ibrutinib plus ofatumumab had active Richter’s transformation (RT) at study entry. Patients were generally high risk, with a median of 3 prior therapies (range, 0-16; 8 patients were previously untreated), 80% with unmutated IGHV, 58% with complex karyotype (3 or more distinct abnormalities), 37% with del(17p13.1), and 26% with del(11q22.3). With a median follow-up of 20 months (range, 2-47 months), 232 patients (75%) remain on therapy, including 7 patients who were removed from the study to pursue stem cell transplant or to receive ibrutinib therapy commercially at another center. Of the 76 patients who discontinued therapy, only 31 (10% of 308 total) discontinued therapy because of disease progression.
Forty-five patients discontinued therapy for reasons other than disease progression. Twenty-eight discontinued because of infection, 8 because of other adverse events, and 9 owing to other events including progressive multifocal leukoencephalopathy in 1 patient who received rituximab 363 days prior, medication noncompliance (n = 1), comorbid medical condition (n = 2), failure to thrive (n = 1), sudden cardiac death (n = 1), need for anticoagulation (n=2), and cerebrovascular event (n = 1). Nonrelapse discontinuations tended to occur early, with only 1 patient discontinuing therapy after 24 months for a reason other than progressive disease. Therapy-limiting infections especially tended to occur early, with a median time to discontinuation of 102 days. Sixteen discontinuations due to infectious toxic effects occurred during the first 6 months of therapy, 7 within 6 to 12 months, and 5 beyond 12 months.
In multivariable modeling, age was the only significant independent risk factor of therapy discontinuation for reason other than disease progression (hazard ratio [HR] for 10-year increase, 1.87; 95% CI, 1.33-2.64 [P < .001]; Table 2). Increasing number of prior therapies was also associated with nonrelapse discontinuation, although it was not statistically significant (HR, 1.09; 95% CI, 1.00-1.19 [P = .054]).
Patients who discontinued ibrutinib therapy for nonrelapse reasons tended to do poorly, with a median survival after discontinuation of 8 days (95% CI, 0-56 days). This was heavily influenced by patients coming off trial because of infection, where 16 of the 28 patients with infection died the same day they came off treatment and only 2 were alive at the last follow-up on days 63 and 176. Conversely, patients who discontinued therapy for noninfectious adverse events or other reasons had a median survival of 238 days, where 7 of the 17 patients had not died at the time of the last follow-up and 4 had survived more than a year following treatment discontinuation (eFigure 1 in the Supplement).
Of 31 patients who discontinued therapy because of disease progression, 13 progressed with CLL and 18 with RT, 3 of whom had previously transformed. Characteristics of these patients are listed in Table 3. While most of the patients with RT developed diffuse large B-cell lymphoma (78%), 1 patient each developed Hodgkin lymphoma, plasmablastic lymphoma, composite B-cell and T-cell lymphoma, and peripheral T-cell lymphoma. Richter’s transformation tended to occur earlier than progressive disease because of CLL, where the estimated cumulative incidence of RT at 12 months was 4.5% (95% CI, 2.0%-7.0%) and the estimated cumulative incidence of progressive disease due to CLL at 12 months was 0.3% (95% CI, 0%-1.0%) (Figure).
Owing to the low frequency of relapse, failures due to RT or CLL were combined in an attempt to identify baseline factors that increased the risk of disease progression. In modeling individual variables associated with disease-related discontinuation and adjusting for monotherapy with ibrutinib, increased number of prior therapies (HR, 1.12; P = .03) BCL6 abnormalities (HR, 3.77; P < .001), MYC abnormalities (HR, 2.59; P = .01), presence of del(17p) (HR, 2.28; P = .03), and complex karyotype (HR, 5.17; P = .003) were all significantly associated with a higher risk of progression (eTable 2 in the Supplement). In multivariable analysis, presence of BCL6 abnormalities (HR, 2.70; 95% CI, 1.25-5.85 [P = .01]) and complex karyotype (HR, 4.47; 95% CI, 1.50-13.34 [P = .007]) remained independent risk factors (Table 2). Increased LDH level at baseline was not significantly associated with disease progression when combining failure types but appeared to be associated with development of RT.
Patients with RT generally did poorly following relapse. Of 18 patients, 6 were unable to receive further therapy and died within 1 month of transformation. One patient was treated at another institution with an unknown regimen. The remaining 11 patients were treated with R-EPOCH (rituximab plus etoposide, prednisolone, vincristine, cyclophosphamide, and doxorubicin) (n = 6), R-CHOP (rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone) (n = 1), R-ICE (rituximab plus ifosfamide, carboplatin, etoposide) (n = 1), rituximab plus dexamethasone (n = 1), OFAR (oxaliplatin, fludarabine, cytarabine, rituximab) (n = 1), or brentuximab (n = 1). One patient was lost to follow-up after treatment was started, but the 10 remaining patients all had disease progression during their next treatment regimen. Median survival from transformation was 3.5 months (95% CI, 10 days to 6.0 months) (eFigure 2 in the Supplement).
Patients with CLL progression tended to have rapid disease progression following discontinuation of ibrutinib therapy. One patient had rapid progression of nodes at another facility and was not seen again at OSU, so details of care are unknown. One patient had primarily progression in ascites and cytopenias and was therefore unmeasurable. Of the remaining 11 patients, 2 had disease progression in the peripheral blood only, 3 had progression in the nodes only, and 6 had both. For patients with peripheral blood disease (n = 8), the median increase in absolute lymphocyte count during the 6 months prior to ibrutinib therapy discontinuation was 17.34 × 109/L (range 6.62-177.34 × 109/L) (to convert to lymphocytes per microliter, divide by .001). All patients had a normal lymphocyte count 6 months prior to therapy discontinuation. For 5 patients where there was a time interval of more than 1 week between discontinuation of ibrutinib therapy and start of next therapy, the mean change in lymphocyte count between relapse and ibrutinib discontinuation was +0.75 × 109/L per day, and was +4.6 × 109/L per day from the time of ibrutinib therapy discontinuation until next therapy. For the patients with nodal disease (n = 9), the median increase in the sum of the products of lymph node diameter of the 3 largest nodes on the computed tomographic scan was 10.85 cm (range, 3.98-104.15 cm). Three patients had palpable lymph nodes and a time of more than 1 week between ibrutinib therapy discontinuation and next therapy. In these patients, the rate of change of the sum of the products of the 3 largest lymph nodes by physical examination was +0.35 cm, 0 cm, and 0.2 cm per day, respectively, prior to ibrutinib therapy discontinuation, and +0.79 cm, 1.18 cm, and 4.63 cm per day after ibrutinib therapy discontinuation.
After disease progression, patients tended to require the initiation of therapy quickly to achieve disease control (Table 3). Of the 13 patients, 11 received further therapy, most within a few weeks of discontinuing ibrutinib therapy (median, 10 days; range, 1-42 days). Two were not able to receive further therapy and died at days 10 and 180, respectively, following discontinuation. Median survival from time of CLL progression was 17.6 months (95% CI, 4.7 months–“not reached”) (eFigure 2 in the Supplement).
We have previously shown that resistance to ibrutinib can be mediated through mutations in BTK or PLCG2 in a small group of patients. To extend these findings, we have performed Ion Torrent deep sequencing for both of these genes at baseline and time of relapse for patients who have relapsed on ibrutinib therapy (Table 4).
In patients who progressed with RT, peripheral blood was obtained at baseline and time of relapse in 8, whereas lymph node biopsy was obtained at relapse from 1 patient. The lymph node at progression showed no mutations in BTK or PLCG2. Peripheral blood in 6 patients at disease progression was negative for mutations, but 2 patients had mutations: 1 had 3 separate mutations in BTK (C481S, T474I, and T474S), and 1 had 3 separate mutations in BTK (C481Y, C481R, and L528W) and D334H mutation in PLCG2. Notably, these 2 patients were the only patients with RT who had a worsening lymphocytosis with CLL phenotype at the same time as RT, so it is possible that they had both progressive CLL as well as RT.
In patients who had CLL progression, samples at baseline (peripheral blood) and at the time of relapse was obtained in 11 of 13 patients (peripheral blood in 9 and bone marrow in 2). Three of these patients have been previously reported19 and had acquired mutations in BTK C481S, PLCG2 R665W, and BTK C481S in association with PLCG2 R665W, L845F, and S707Y. In the 8 remaining patients, 5 had sole mutations in BTK with C481S, 1 had BTK C481F mutation, and 1 had 5 separate mutations in PLCG2 (R665W, S707P, S707F, R742P, and L845fs). One additional patient had both BTK and PLCG2 mutations, with BTK C481S and PLCG2 D1140G. Thus, all patients who have been identified at our institution to have progressive CLL after ibrutinib therapy have had mutations in either BTK or PLCG2 that were acquired during therapy. Relapse mutations were not identified before treatment in any patient with deep sequencing (Table 4).
We describe herein the first large single-center experience of patients treated with ibrutinib, with a median follow-up of almost 2 years. We show that nonrelapse-related discontinuation of therapy is more common early in therapy and becomes progressively infrequent. We show that RT and CLL progressions are both relatively uncommon in this very high-risk population, with RT tending to occur earlier and CLL progression later. Finally, we confirm and extend our previous findings that CLL progression is associated with mutations in BTK or PLCG2.
While previous publications of clinical trials report reasons for discontinuation of ibrutinib therapy, this report greatly extends these findings by including a larger number of patients than any individual trial and longer follow-up than other reports. Also, we report for the first time to our knowledge a multivariable analysis identifying factors at baseline that are associated with discontinuation and outcomes following ibrutinib therapy discontinuation, which will be relevant for general oncology practice.
Perhaps most significant, these data confirm that ibrutinib is a drug capable of shifting the paradigm of therapy for relapsed disease. In our group of high-risk patients, the estimated cumulative incidence of disease progression at 18 months was only 8.9% (95% CI, 5.2%-12.6%). As more patients are treated with ibrutinib off study, we expect that outcomes will not be as positive as in clinical trials, given variability in compliance and follow-up, but we should see patients with relapsed CLL living longer than has been seen in the past. As our data show an estimated cumulative incidence of nonrelapse discontinuation of 15.6% (95% CI, 11.1%-20.0%) at 18 months, there remains a need to identify new therapies for patients who do not tolerate ibrutinib or who are unable to continue because of concomitant medications.
These data also confirm and extend earlier reports of the association of BTK and PLCG2 mutations with CLL relapse. Interestingly, in our experience thus far, these mutations are associated with all CLL relapses. While we have not yet characterized all of these mutations, our research thus far has shown that BTK C481 mutations interfere with ibrutinib binding, and PLCG2 mutations potentially have gain-of-function effects. We are currently characterizing the additional mutations identified as part of this analysis. From this small subset of patients, it appears that these mutations may help to identify a patient who is relapsing with CLL. Importantly, we still do not see evidence of these mutations at baseline, suggesting that they are indeed the result of selective pressure from the drug, although sampling only peripheral blood compartment is a limiting factor, and it is possible that these mutations may be present in the blood at very low levels or in other niches at baseline. Finally, our data suggest that RT is not associated with these mutations in most cases and opens the question of resistance mechanisms in this patient population.
While this analysis presents data from a large number of patients and we attempt to control for confounding variables through multivariable analyses, our study is limited by the retrospective nature of the data and the single-center experience. It is possible that patterns at our institution may not be generalizable to the entire ibrutinib-treated population, and it will be important to confirm these findings in larger multicenter patient populations. However, given that this drug is relatively new and discontinuations are infrequent with short follow-up, it will likely be many years before larger series with longer follow-up times are available.
Ultimately, we show that patients with disease progression while on ibrutinib therapy, especially those with RT, have poor outcomes. While the incidence of disease progression is low with this duration of follow-up, it is expected that more patients will experience relapse with extended follow-up. We show that disease tends to progress quickly in patients, especially when the drug therapy is stopped, which points to a need for clinical trials to allow shorter washout periods for these patients. Also, physicians may consider waiting before discontinuing ibrutinib therapy until an alternative therapeutic plan can be formed and potentially even continuing ibrutinib therapy in combination with RT-directed therapy in select patients.
These data show that continued clinical trials are needed in this disease and that scientific progress cannot stop even with a breakthrough drug like ibrutinib. Patients with RT remain a high research priority to identify new targets and new therapies. Also, the question of whether high-risk patients may benefit more from transplant or kinase inhibitor is unanswered. There remains much to be learned about the biological mechanisms of CLL from the use of kinase inhibitor drugs, and the ideal drug, target, or combination for all patients is likely yet to be identified.
Accepted for Publication: December 11, 2014.
Corresponding Author: Jennifer A. Woyach, MD, Division of Hematology, Department of Internal Medicine, The Ohio State University, 455A OSU CCC, 410 W 12th Ave, Columbus, OH 43210 (Jennifer.Woyach@osumc.edu).
Published Online: February 26, 2015. doi:10.1001/jamaoncol.2014.218.
Author Contributions: Dr Woyach had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Drs Maddocks and Lozanski and Ms Ruppert contributed equally to this work.
Study concept and design: Maddocks, Blum, Grever, Johnson, Byrd, Woyach.
Acquisition, analysis, or interpretation of data: Maddocks, Ruppert, Lozanski, Heerema, Zhao, Abruzzo, Lozanski, Davis, Gordon, Smith, Mantel, Jones, Flynn, Jaglowski, Andritsos, Awan, Blum, Byrd, Woyach.
Drafting of the manuscript: Maddocks, Ruppert, Gordon, Mantel, Byrd, Woyach.
Critical revision of the manuscript for important intellectual content: Maddocks, Ruppert, Lozanski, Heerema, Zhao, Abruzzo, Lozanski, Davis, Smith, Jones, Flynn, Jaglowski, Andritsos, Awan, Blum, Grever, Johnson, Byrd, Woyach.
Statistical analysis: Ruppert.
Obtained funding: Blum, Johnson, Byrd, Woyach.
Administrative, technical, or material support: Maddocks, Lozanski, Zhao, Abruzzo, Lozanski, Davis, Gordon, Mantel, Flynn, Andritsos, Awan, Blum, Grever, Byrd, Woyach.
Study supervision: Maddocks, Blum, Byrd, Woyach.
Conflict of Interest Disclosures: Dr Maddocks has received research support and has served as a consultant for Pharmacyclics. Dr Blum has received research funding from Pharmacyclics and Janssen. Dr Jaglowski has received research funding from Pharmacyclics. Dr Jones has served as a consultant for Pharmacyclics and Janssen. No other disclosures are reported.
Funding/Support: This work was supported by a Specialized Center of Research from the Leukemia and Lymphoma Society; grants P50-CA140158, P01 CA95426, R01 CA177292, and K23 CA178183 from the National Cancer Institute; The D. Warren Brown Foundation; Four Winds Foundation; The Sullivan Chronic Lymphocytic Leukemia Research Fund; Mr and Mrs Michael Thomas; Mr and Mrs Al Lipkin; The Harry T. Mangurian Jr Foundation; Pelotonia; and Conquer Cancer Foundation.
Role of the Funder/Sponsor: Pharmacyclics was involved in the design and conduct of OSU 10032, OSU 10053, and OSU 12024. Pharmacyclics provided study drug for OSU 11133. For this article, Pharmacyclics had no role in the collection, management, analysis, and interpretation of the data; preparation or approval of the manuscript; and decision to submit the manuscript for publication. Pharmacyclics did review the manuscript prior to submission.
Additional Contributions: The authors wish to thank the patients and families who provided samples for this work.