By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Wiley Online Library will be unavailable on Saturday 7th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 08.00 EDT / 13.00 BST / 17:30 IST / 20.00 SGT and Sunday 8th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 06.00 EDT / 11.00 BST / 15:30 IST / 18.00 SGT for essential maintenance. Apologies for the inconvenience.
Although rituximab-based chemoimmunotherapy (CIT) has substantially improved clinical outcomes in chronic lymphocytic leukemia (CLL), only 40% to 50% of patients achieve a complete remission (CR). There remains interest in identifying new approaches to improve the effectiveness of CIT. Ofatumumab is a fully human anti-CD20 monoclonal antibody with greater apparent single-agent activity than rituximab in CLL patients.
Previously untreated CLL patients in need of therapy received 6 cycles of CIT induction with pentostatin, cyclophosphamide, and ofatumumab (PCO) followed by response assessment.
Of the 48 patients enrolled, 77% completed PCO induction. Adverse events during induction included grade 3+ hematologic toxicity (27%) and grade 3+ nonhematologic toxicity (23%). Median CD4 count after induction and 6 months later were 186 × 106/L and 272 × 106/L. The overall response rate was 96% (46 of 48 patients), and the CR rate was 46% (22 of 48 patients). Among the 38 patients who underwent minimal residual disease evaluation, 7 (18%) were negative for minimal residual disease. After median follow-up of 24 months, 10 (21%) patients have progressed and 8 (17%) have required retreatment. The efficacy and toxicity of ofatumumab-based CIT compare favorably to our historical trials of rituximab-based CIT using an identical chemotherapy backbone (n = 64). Time to retreatment also appeared longer for ofatumumab-based CIT (free of retreatment at 24 months: 86% [95% confidence interval = 75-99] versus 68% [95% confidence interval = 56-81] for rituximab-based CIT).
The last decade has been a time of tremendous progress in the treatment of chronic lymphocytic leukemia (CLL). Phase 3 trials conducted in the 1990s established that fludarabine-based therapy offered superior response rates and progression-free survival compared with alkylating agent–based therapy.[1-3] Randomized controlled trials subsequently demonstrated that the combination of fludarabine and cyclophosphamide (FC) further improved response rates and progression-free survival.[4-6] These advances were followed by the development of chemoimmunotherapy (CIT) combining anti-CD20 monoclonal antibodies (mAbs) with purine nucleoside analogues. A seminal phase 3 trial conducted by the German CLL Study Group demonstrated that the addition of the anti-CD20 mAb rituximab to the FC platform (FCR) improved not only response rates and progression-free survival but also overall survival. Unlike previous treatment approaches, a large proportion of patients treated with CIT have no detectable disease at the completion of induction therapy even when evaluated using highly sensitive assays (minimal residual disease [MRD] negativity). MRD status at the completion of therapy has been found to be a strong predictor of both progression-free and overall survival.
Despite these advances, only ∼45% of patients achieve a complete remission with CIT such as FCR, and nearly all patients eventually relapse. Furthermore, FCR showed considerable hematologic toxicity, particularly among patients greater than age 70 years. These facts have generated interest in strategies to improve both the efficacy and tolerability of CIT. All initial CIT trials for CLL employed rituximab as the anti-CD20 mAb. Rituximab is a chimeric mAb with a human constant region and a murine variable region which appears to mediate its effects in CLL primarily through antibody-dependent cell-mediated cytotoxicity (ADCC) rather than through either complement fixation or direct induction of apoptosis. In efforts to improve clinical efficacy, more recently developed anti-CD20 mAbs have been engineered to invoke greater ADCC, induce direct cell death, or promote complement-dependent cytotoxicity (CDC). Ofatumumab is a fully human anti-CD20 mAb with clinical activity as a single agent in patients with fludarabine-refractory CLL. Ofatumumab binds to both the small and large loop domains of CD20 at an epitope distinct from rituximab and is more efficient at inducing CDC with similar cell-mediated cytotoxicity. Ofatumumab appears to have greater single-agent clinical activity than rituximab[10, 12, 13] in patients with previously treated CLL and also has activity in rituximab-refractory patients. Given these distinct characteristics of ofatumumab, we conducted a trial of ofatumumab-based CIT for patients with previously untreated CLL and compared the results to our historical trials of rituximab-based CIT.
MATERIALS AND METHODS
Eligible patients were required to have a diagnosis of progressive CLL in need of treatment according to the National Cancer Institute/International Workshop on Chronic Lymphocytic Leukemia (NCI/IWCLL) Working Group criteria. Patients had to be previously untreated, have an ECOG performance status of 0 to 2, and have adequate renal and hepatic function. Patients with recent (< 1 month) myocardial infarction, class III or IV heart failure, uncontrolled infection, infection with human immunodeficiency virus (HIV), active hepatitis B or C infection, or active hemolytic anemia were excluded. Patients with other malignancies were allowed to participate, provided they were not receiving treatment and had a life expectancy > 2 years. There was no upper age limit on eligibility. The protocol was reviewed and approved by both the Mayo Clinic and Duke University institutional review boards and was registered with clinicaltrials.gov (identifier NCT01024010).
After giving written informed consent, eligible patients with CLL received 6 cycles of induction CIT using pentostatin (2 mg/m2 on day 1), cyclophosphamide (600 mg/m2 on day 1), and ofatumumab (cycle 1: 300 mg on day 1, 1000 mg/m2 on day 2; cycles 2-6: 1000 mg/m2 on day 1) given intravenously every 21 days. Methylprednisolone 80 mg (or equivalent) was administered intravenously prior to each dose of ofatumumab.
Prophylaxis against Pneumocystis jiroveci (sulfamethoxazole-trimethoprim or alternative) and herpes zoster (valacyclovir or alternative) were given to all patients for 1 year from the start of cycle 1. All patients were given allopurinol (300 mg orally once daily) on days 1 through 14 of cycle 1. Pegfilgrastim was administered on day 2 of each cycle.
Patients completing 6 cycles of PCO induction underwent complete restaging including evaluation for MRD using flow cytometry (assay sensitivity to at least 10−4 [< 0.01%]). Restaging occurred 12 weeks after day 1 of cycle 6. MRD assays after induction were preferentially performed on bone marrow aspirate (n = 36) with use of peripheral blood if bone marrow aspirate was not available (n = 2).
Dose Modifications for Toxicity
Platelet and hemoglobin adverse events were graded according to the IWCLL CLL Working Group grading scale for hematologic toxicity. All other adverse events were graded according to the NCI Common Toxicity Criteria (version 4).
Criteria for Response
Responses were graded according to the NCI/IWCLL Working Group criteria. Bone marrow biopsies were performed at registration and the response evaluation following PCO induction to document complete response. Computed tomography (CT) scans of the chest, abdomen, and pelvis were performed in all patients at registration and at the response evaluation following PCO induction.
Quantification of T Cell Numbers and Immunoglobulin G Levels at Baseline and During Treatment
Immunoglobulin G (IgG) levels and the absolute number of CD4+ and CD8+ T cells were measured using routine clinical assays.
A 1-stage binomial design was used in this phase 2 trial. The primary endpoint was the proportion of patients who achieved a complete response (CR). The study was initially designed to accrue a total of 30 evaluable patients to test the null hypothesis that the true CR rate is at most 25% versus the alternative hypothesis that it is at least 50%. The study had 90% power, with a 5% type I error rate. The sample size was later increased to 48 patients. A patient was considered evaluable for response if they were eligible and initiated treatment. The proportion of CRs was estimated by the number of patients who achieved a CR divided by the total number of evaluable patients. Time to retreatment (TTR) was defined as the time from registration to initiation of subsequent treatment for CLL. The distribution of time to retreatment was estimated using the Kaplan-Meier method. Laboratory values measured on a continuous scale (eg, IgG levels, T-cell counts) were compared between paired sample time points using the Wilcoxon signed-rank test. Clinical characteristics and prognostic factors were compared to a historical cohort using Fisher's exact test and Wilcoxon's rank-sum test for categorical and continuous factors, respectively.
Patient Characteristics and Response to Induction
Forty-eight patients were enrolled between March 2010 and March 2011, all of whom were eligible for treatment. The median age was 65 years (range, 50-83 years) and 71% of patients were male (Table 1); 35% of patients had intermediate Rai risk and 65% high Rai risk. On prognostic testing, 48% CD38+, 48% were ZAP-70+, and 48% were IGHV (immunoglobulin heavy variable cluster) unmutated (um). Chromosomal analysis by fluorescence in situ hybridization (FISH) found that 69%,10%, and 4% of patients had 1, 2, or 3 FISH defects, respectively. FISH risk category using the classification of Dohner et al is shown in Table 1.
All patients have completed active treatment. Patients received a median of 6 cycles (range, 2-6 cycles) with 37 of 48 (77%) patients completing the intended 6 cycles of therapy. Eleven patients discontinued therapy early due to adverse events/complications (9 patients), disease progression (1 patient), and death while on study (1 patient). The adverse events/complications necessitating discontinuation of treatment included rash (2 patients), autoimmune hemolytic anemia (2 patients), decline in functional status (1 patient), infection (1 patient), anorexia/fatigue (2 patients), and repeated hospitalizations (1 patient). While receiving treatment, 12 of 48 (25%) patients experienced a dose delay and 2 of 48 (4%) required dose reductions. Adverse events deemed at least possibly related to PCO are shown in Table 2 and included 13 (27%) patients with grade 3+ hematologic toxicity and 11 (23%) with grade 3+ nonhematologic toxicity. The most common grade 3 or 4 toxicity was neutropenia, which occurred in 11 (23%) patients. Four (8%) patients experienced grade 3 or higher thrombocytopenia. Patients aged 70 years or older appeared more likely to experience grade 3+ hematologic toxicity, although this difference did not reach statistical significance (age < 70: 20% versus age ≥ 70: 39%; P = .19). No clear difference in grade 3+ nonhematologic toxicity was observed with age (age < 70: 20% versus age ≥ 70: 28%; P = .72).
Table 2. Grade 3 and 4 Adverse Events Attributed to PCO Therapy
Abbreviations: CTCAE, Common Terminology Criteria for Adverse Events; PCO, pentostatin/cyclophosphamide/ofatumumab.
International Workshop on Chronic Lymphocytic Leukemia Grading Scale for hematological toxicity.
Changes in IgG levels, CD4 cells, and CD8 cells with therapy are shown in Figure 1. The median CD4 cell count decreased from 1304 × 106/L at baseline to 186 × 106/L after PCO induction (n = 38; P = .0001). Among 29 patients with CD4 cell counts measured 6 months after the completion of induction, 19 of 29 patients (66%) had a CD4 count > 200 × 106/L (median, 272 × 106/L).
Response to Therapy
All patients were evaluable for response. The overall response rate (ORR) to PCO by the IWCLL criteria was 96% (95% confidence interval [CI] = 86%-99%) with 22 (46%) CR/CR with incomplete marrow recovery (CRi), 3 (6%) complete clinical response (CCR; who went off study early and did not have bone biopsy performed), and 21 (44%) partial remission (PR; including 14 [29%] nodular PR [nPR]). Among the 2 patients with deletion 17p13, 1 patient had PR and 1 patient had no response. Thirty-eight of 46 (83%) responding patients underwent evaluation for MRD by sensitive flow cytometry. Among these, 7 of 38 (18%) were MRD-negative, including 5 of 19 (26%) patients with CR/CRi who underwent MRD evaluation of the bone marrow.
After treatment, 10 of 39 (26%) patients who had no palpable lymphadenopathy or organomegaly and who underwent CT scan had residual lymphadenopathy (defined as > 1.5 cm) or organomegaly as determined by imaging. Considering the results of CT scan in response assessment would reclassify 6 of 22 (27%) patients with CR/CRi to PR, including 1 of 5 patients with MRD-negative CR by conventional criteria. CT scan findings would also reclassify 2 of 14 patients with nPR (a subcategory of PR) to standard PR.
As of last follow-up, 10 patients have progressed and 5 patients have died. No patients were lost to follow-up. Deaths were due to disease progression (n = 2), pneumonia (n = 1), respiratory failure (n = 1), and unknown cause (n = 1). The median follow-up of living patients is 24 months (range, 15-34 months). To date, 8 of the 46 (17%) responding patients have developed disease progression, and 6 (13%) have required retreatment. Median TTR has not yet been reached (NR) (95% CI = 30 months to NR). The median TTR for all patients is shown in Figure 2.
Comparison of response was performed by IGHV mutation, CD38, CD49d, and ZAP-70 status as well as by the results of FISH analysis (high risk: del[17p13.1] or del[11q22.3]; low risk/intermediate risk: del[13q13], trisomy 12, or negative). Depth of response (CR/CRi versus CCR/nPR/PR) was not correlated with any pretreatment prognostic factor.
Comparison to Historical Cohort of Rituximab-Based CIT
Finally, we compared this cohort of patients treated with ofatumumab-based CIT to our prior cohort of patients treated with rituximab-based CIT using an identical chemotherapy backbone (pentostatin and cyclophosphamide). It should be emphasized that this analysis represents a comparison of patients treated on sequential phase 2 trials rather than being a randomized comparison. Nonetheless, the eligibility criteria were nearly identical between the trials, and the majority of patients accrued at the same academic center. To compare toxicity and response of ofatumumab-based CIT to rituximab-based CIT, the 48 patients in the current study were compared to 108 patients treated with PCR (pentostatin, cyclophosphamide, and rituximab) induction (before any consolidation was administered). To compare the TTR, all patients in the current trial were compared to all 64 patients in the original PCR trial. Patients in our trial of PCR followed by lenalidomide consolidation were not included in the TTR analysis because they received consolidation therapy.
The clinical characteristics and prognostic profile of patients treated with ofatumumab-based CIT and rituximab-based CIT are shown in Table 3. The age, sex, absolute lymphocyte count, beta-2-microglobulin levels, ZAP-70 status, and FISH profile of treated patients were similar (all P > .05). The ofatumumab cohort included more advanced stage patients (Rai stage III/IV: 65% versus 47%; P = .06) and could be considered slightly higher risk on that basis. Although not statistically significant, the ofatumumab cohort also had a slightly higher rate of CD38+ patients (48% versus 33%; P = .11) and a slightly lower rate of IGHV UM patients (49% versus 63%; P = .11).
Table 3. Comparison of Patient Characteristics: Ofatumumab-Based CIT to Rituximab-Based CIT
Current Cohort PCO (N = 48)
Historical Cohort Trial PCR Without Consolidation (N = 108)
With respect to hematologic toxicity during treatment, ofatumumab-based CIT compared favorably to our previous experience with rituximab-based CIT (≥ grade 3 hematologic toxicity ofatumumab-CIT = 27% [13 of 48] versus rituximab-CIT = 41% [44 of 108], P = .11). Grade ≥ 3 neutropenia was experienced by 11 of 48 (23%) patients treated with ofatumumab-CIT as compared to 37 of 108 (34%) patients treated with rituximab-CIT (P = .19). No difference in nonhematologic toxicity was observed during treatment (≥ grade 3 nonhematologic toxicity ofatumumab-CIT = 19% [9 of 48] versus rituximab-CIT = 24% [26 of 108], P = .54). Despite a higher proportion of Rai stage III/IV patients in the ofatumumab cohort, the efficacy of ofatumumab-based CIT also compared favorably to our previous experience with rituximab-based CIT (Table 4). Given the nonsignificant but slightly higher number of IGHV UM patients in the rituximab-CIT cohort, we also evaluated CR rates in IGHV UM patients. Numerically higher CR rates for ofatumumab-CIT persisted on this subanalysis of IGHV UM patients (CR rate ofatumumab-CIT: 52% [12 of 23]; CR rate rituximab-CIT = 38% [25 of 66]).
Table 4. Comparison of Response to Therapy: Ofatumumab-Based CIT to Rituximab-Based CIT
With the caveat that it represents a comparison across sequential phase 2 trials, the proportion of patients free of retreatment at 24 months was 86% (95% CI = 75%-99%) for ofatumumab-based CIT versus 68% (95% CI = 56%-81%) for rituximab-based CIT (Fig. 3).
Although the addition of the anti-CD20 mAb rituximab to purine analogue–based chemotherapy improves overall survival in patients with CLL, it remains a noncurative treatment strategy. The considerable single-agent activity of ofatumumab and its ability to promote CDC has raised hope that ofatumumab-based CIT may improve efficacy. In the present study, ofatumumab-based CIT was well tolerated and had high ORR and CR rates. The 23% prevalence of grade 3+ neutropenia compares favorably to our previous experience with rituximab-based CIT (34%),[18, 19] using the same pentostatin and cyclophosphamide platform, as well as with prior studies of FCR CIT which have reported grade 3+ neutropenia in 34% of patients. MRD-negative remissions in the bone marrow were achieved in 7 patients (18% of those tested). Although this rate of MRD-negative remissions in the bone marrow appears lower than the 44% reported with FCR, the 2-year freedom from retreatment (86%) compares favorably to the ∼80% level observed in historical studies of FCR and the ∼70% level reported in studies of fludarabine and rituximab (FR).[20, 21]
This is the second report of ofatumumab-based CIT for patients with previously untreated CLL. Wierda et al reported the results of an international phase 2 trial of ofatumumab in combination with fludarabine and cyclophosphamide (FCO) in 61 previously untreated patients with CLL. Ofatumumab was administered at 1 of 2 dose levels (dose level 1: 500 mg/cycle; dose level 2: 1000 mg/cycle). The rate of grade 3+ neutropenia with FCO was 35% for dose level 1 and a striking 60% for dose level 2 (the ofatumumab dose used in our study). Even though the median age in the FCO trial was nearly a decade younger (median age, 56 years) than the patients in our study, only 64% of FCO-treated patients completed the intended 6 cycles of therapy (57% of patients on dose level 2) with prolonged cytopenias the primary reason for patients to withdraw. The ORRs observed with the FCO combination were disappointing (ORR dose level 1: 77%; ORR dose level 2: 73%) relative to historical trials of rituximab-based CIT using the fludarabine, cyclophosphamide (FC) platform[7, 23] but increased to ∼90% range when patients who experienced CR with incomplete marrow recovery were considered. The CR rates observed with FCO (dose level 1: 32%; dose level 2: 50%) were difficult to interpret due to the lack of a well-matched historical cohort for comparison. These CR rates are dramatically lower than those of single-center phase 2 trials of FCR, but are similar to previous international, multicenter trials. Compared to our historical experience with rituximab-based CIT using the pentostatin and cyclophosphamide platform in a very similar patient population, the ORR (ofatumumab-based CIT: 96%; rituximab-based CIT: 92%) and CR (ofatumumab-based CIT: 46%; rituximab-based CIT: 37%) observed with ofatumumab-based CIT compare favorably to that of rituximab-based CIT. Ofatumumab-based CIT also appeared to prolong TTR relative to our historical experience with rituximab-based CIT. Collectively, these data suggest a randomized comparison of rituximab-based CIT to ofatumumab-based CIT in CLL may be warranted.
The restriction of CD20 glycoprotein expression to B cells has made CD20 an attractive therapeutic target for B-cell malignancies. The precise mechanism of action of anti-CD20 mAb in CLL remains controversial and likely differs among the different anti-CD20 antibodies. Rituximab is thought to primarily act through ADCC rather than CDC or direct induction of cell death.[25, 26] Ofatumumab was engineered to bind a different epitope of CD20 that is in close proximity to the cell membrane and has been found to induce substantial CDC. Obinutuzumab (GA101) is another anti-CD20 antibody in clinical testing that is a humanized type II anti-CD20 mAb thought to mediate its effect by direct killing and cell-mediated cytotoxicity. Recent studies compared the effects of these 3 anti-CD20 mAbs on CLL B cells in vitro. These studies suggest that ofatumumab induces the greatest complement activation and antibody-dependent phagocytosis by macrophages among the 3 antibodies, whereas obinutuzumab induces the greatest direct cytotoxicity (without in vitro crosslinking) and natural killer cell–mediated ADCC. The clinical relevance of these findings is unknown and which anti-CD20 mAb has the best efficacy as a single agent and in combination with chemotherapy can only be determined through randomized controlled trials.
The response criteria for CLL continue to be based on the findings of physical examination and bone marrow biopsy at the completion of treatment. Among the 39 patients in our study with no palpable lymphadenopathy or organomegaly by examination, CT scanning at the time of response evaluation identified residual lymphadenopathy in 26% including 6 of 22 (27%) of patients who would be classified as CR based on the results of physical examination and bone marrow biopsy. Residual lymphadenopathy on CT scans was also detected in 1 out of every 5 patients with MRD-negative CR. These results are consistent with previous studies that suggest residual disease by CT scan is detectable in ∼20% to 30% of patients with a CR by the NCI/IWCLL response criteria.[5, 29, 30] Follow-up is too short and too few events have occurred to determine whether patients in our study with no palpable lymphadenopathy by examination but with residual abnormalities on CT have shorter TTR. Currently, the best available evidence suggests that disease detectable by CT scan in patients meeting the NCI/IWCLL criteria for CR does not impact progression-free survival in patients treated with CIT. The role of CT imaging in the CLL the response evaluation may become more important if effective consolidation/maintenance strategies are developed.
Despite our encouraging results with ofatumumab-based CIT, several qualifications and important questions remain. First, the comparisons of efficacy and toxicity of ofatumumab-based CIT relative to rituximab-based CIT are derived from comparison to a historical cohort. Although patients met nearly identical eligibility criteria and the age, sex, absolute lymphocyte count, and prognostic profile of treated patients in the 2 cohorts were similar, the suggested improvement in TTR provides a level of evidence far below a randomized trial. Second, because this study opened ∼3 years ago and accrued rapidly, the current median follow-up (24 months) is relatively short. Although > 80% of patients have not required retreatment at the 24-month time point, longer follow-up will be needed to assess the durability of these remissions. Third, although ∼40% of the patients in our trial were ≥ age 70 years and PCO therapy appeared well tolerated among the older patients in this study, the median age of study participants remains younger than patients with CLL in the general population.
The results of this phase 2 study suggest ofatumumab-based CIT is well tolerated in patients with previously untreated CLL. The efficacy and toxicity profile of ofatumumab-based CIT compares favorably to historical trials of rituximab-based CIT, suggesting randomized trials comparing ofatumumab-based CIT and rituximab-based CIT should be considered.
This clinical trial and the associated correlative studies were funded through a grant from GlaxoSmithKline (GSK).
CONFLICT OF INTEREST DISCLOSURE
Dr Shanafelt has received funding for Mayo Clinic research from Celgene, Hospira, Genentech, GSK, Cephalon, and Polyphenon E International. Dr Lanasa has been a consultant for Genetech, Hospira, and had a consulting relationship with GSK that was terminated prior to study initiation. Dr Beaven has received funding for Duke University research from Celgene, Novartis, and Spectrum, and has a family member who works for GSK. Dr Kay has received funding for Mayo Clinic research from Celgene, Genentech, Gilead, Pharmacyclics, and has been a consultant for Janssen Research and Development, Novartis. Dr Zent has received funding for Mayo Clinic research from Genentech, Genzyme, Novartis, GSK, and Biothera. All other authors made no disclosure.