ASCO 2017 | CLL: results of the phase II CALGB 10404 and phase III GENUINE trials, and a retrospective analysis of Richter’s Transformation with novel agents 

At the 2017 American Society of Clinical Oncology (ASCO) Annual Meeting, an oral abstract session took place that was jointly chaired by John M. Pagel, MD, PhD, of the Swedish Cancer Institute, Seattle, WA, USA, and Ranjana H. Advani, MD, from the Stanford Cancer Institute, Stanford, CA, USA.

Three key oral abstracts were presented on Chronic Lymphocytic Lymphoma (CLL).

Abstract 7503

The first was presented by Amy S. Ruppert, MAS, from The Ohio State University, Columbus, OH, USA, and reported the results of the Cancer and Leukemia Group B (CALGB) 10404 (Alliance) trial.

The abstract began by explaining that, before the discovery and use of targeted agents for the treatment of CLL such as BTK inhibitors (e.g. ibrutinib), PI3K inhibitors (e.g. idelalisib), and BCL-2 inhibitors (e.g. venetoclax), there was much disagreement concerning which chemoimmunotherapy regimen was the best to use as a backbone for therapy in patients who did not harbor del(11q). Moreover, the impact of the immunomodulatory agent lenalidomide as treatment in these patients was unknown.

Previously, it has been reported that high CR and ORRs can be achieved with fludarabine, cyclophosphamide, and rituximab (FCR; Keating et al. 2005). However, recently this regimen has been associated with short-term toxicity and late onset of treatment-related myeloid neoplasia (Thompson et al. 2016; Fischer et al. 2016). It has been suggested in numerous reports (Eichhorst et al. 2006; Tsimberidou et al. 2008; Hallek et al. 2010) that treatment with cyclophosphamide may benefit patients harboring del(11q). Furthermore, fludarabine plus rituximab (FR) without cyclophosphamide has demonstrated high response rates and durable remissions in patients with treatment naïve, symptomatic CLL (Byrd et al. 2003; Woyach et al. 2011). Lenalidomide has also been reported to have clinical activity in both treatment naïve and R/R patients with CLL, and it has been suggested that high-risk patients harboring del(11q) may benefit from lenalidomide (Ferrajoli et al. 2008). However, caution is required, as high doses of lenalidomide have been associated with tumor flare in CLL patients (Andritsos et al. 2008).

The CALGB 10404 trial (NCT00602459) was a phase II, genetic risk-stratified, randomized trial which aimed to assess different combinations of fludarabine and the anti-CD20 antibody rituximab in patients with symptomatic, newly diagnosed CLL. In patients without del(11q), FR, FR plus 6 months of consolidative lenalidomide (FR+L), and FCR regimens were investigated. FCR and FCR+L were also investigated in patients harboring del(11q). Watch our video interview with Professor Susan O’Brien about this study and hear her take on the del(11q) data.

The primary outcome measure of the trial was 2-year PFS. Overall, 342 non-del(11q) patients were randomly assigned either to the FR (arm A; n=123), FR+L (arm B; n=109), or FCR (arm C; n=110) groups and baseline characteristics were balanced. In non-del(11q) patients, the median age of arms A, B, and C were 61 (28–81), 62 (36–79), and 60 (32–80) years, respectively.

In non-del(11q) patients, 2-year PFS with FR was 64% (90% CI, 57–71%), compared to 71% (90% CI, 63–78%) in FR+L patients, and 74% (66–80%) in the FCR group. Moreover, median PFS was substantially reduced in patients receiving FR (43 months; 95% CI, 33­–50) vs. FR+L (66 months; 95% CI, 45–not reached; P = 0.03) or FCR (78 months; 95% CI, 58–not reached; P < 0.01). In multivariate analysis, a 2-fold increase in WBC count was associated with PFS response (P = 0.002). In del(11q) patients in arm C, 2-year PFS was 56% and those who received FCR+L (arm D) was 65%.

Best responses in non-del(11q) patients were:

• FR: 32% CR, 43% PR
• FR+L: 33% CR, 36% PR
• FCR: 51% CR, 20% PR

Best responses in del(11q) patients were:

• FCR (arm C): 19% CR, 41% PR
• FCR+L (arm D): 35% CR, 39% PR

For all arms, median OS has not been reached. Additionally, 1-year, 2-year, and 3-year OS are similar between arms; however, a plateau in OS with no events beyond 49 months has been observed in the FR+L group. Events continue to occur in the FR and FCR groups.

*1 grade 5 autoimmune hemolytic anemia reported in patient receiving FR induction; 5 grade 5 events (JC virus encephalopathy in arm A, disease progression in arm B, and pneumonitis, sepsis, sepsis/diarrhea in arm C)

This abstract was concluded by stating that patients who received FR+L and FCR treatment achieved superior PFS versus those who received FR. The addition of brief consolidation with lenalidomide to FR demonstrated acceptable toxicity and improved responses compared to FR alone. A plateau in survival distinguishes FR+L treatment regimen from FR alone and FCR; this warrants further investigation. Lastly, treatment based on central interphase cytogenetics to identify high-risk characteristics, such as del(11q), is a feasible strategy.

Abstract 7504

Next, results from the phase III GENUINE trial, the first randomized trial to evaluate the combination of ibrutinib and a novel agent, were presented by Jeff Porter Sharman, MD, from the Willamette Valley Cancer Institute and Research Center/US Oncology Research, Eugene, OR, USA.

Patients with high-risk characteristics such as mutated TP53 or del(11q) have been reported to have abysmal outcomes with single-agent ibrutinib. (O’Brien et al. ASH 2016. Abstract 233). Relapse in high-risk patients remains a challenge and represents an area of urgent unmet medical need. One strategy to combat this is to improve outcomes with ibrutinib by combining it with other agents.

The GENUINE trial (NCT02301156) aimed to determine the efficacy and safety of ibrutinib combined with ublituximab versus ibrutinib monotherapy in R/R high-risk CLL patients. Ublituximab is a chimeric, recombinant IgG1 monoclonal antibody, which target CD20 and induces B-cell-directed Complement Dependent Cytotoxicity (CDC) as well as Antibody-Dependent Cellular Cytotoxicity (ADCC) resulting in B-cell apoptosis. Previous reports indicate that ublituximab is active in CD20+ patients. In fact, in a phase II study of ublituximab combined with ibrutinib an investigator assessed ORR of approximately 88% was reported (Sharman et al. 2016).   

To be eligible, patients must have been found to have at least one high-risk cytogenetic feature defined by the presence of del(17p), del(11q), and/or TP53 mutation. Patients who had previous exposure to ibrutinib were excluded. Patients from the US and Israel were randomized at a ratio of 1:1 to receive either single-agent ibrutinib (IB arm) at 420mg QD or combined with ublituximab (IB+UTX arm; 900mg on days 1, 8, and 15 of cycle 1, then day 1 of cycles 2–6, and then Q3 cycles from then on). ORR (as per iwCLL 2008 criteria) and PFS were the co-primary outcome measures; however, due to the lower than planned enrollment, the study was no longer powered to detect PFS. After a protocol amendment, PFS was used as a secondary outcome measure as well as CR rate, MRD negativity, Time to Response (TTR), and safety.

Of 126 randomized patients, 117 underwent treatment (59 on IB + UTX arm; 58 on IB arm); some dropouts were attributed to ibrutinib being a commercial supply, resulting in access problems. The median age of patients treated was 67 years and patients had received a median of 3 previous treatments (range, 1–8). Between arms, high-risk cytogenetics were similar: approximately 50% of patients in each arm harbored del(17p).

IB + UTX was well tolerated, with infusion related reactions being the most common adverse event (all grades 54%; grade 3–4 5%). A slightly higher rate of all-grade neutropenia was observed in patients receiving IB + UTX (22%) vs. IB monotherapy (7%); however, rates of grade 3–4 neutropenia were similar between arms (9% vs. 10%, respectively). Other adverse events occurred at similar or lower rates with IB + UTX vs. IB alone (all grades), such as fatigue (27% vs. 33%), dizziness (15% vs. 22%), and contusion (15% vs. 29%), anemia (14% vs. 17%).

After a median follow-up of 11.4 months, ORR for IB + UTX was 78% vs. 45% in IB patients, all CRs occurred in the IB + UTX arm (7%). When including PR with lymphocytosis rates (5% vs. 14%), ORR was 83% vs. 59% in the IB + UTX and IB alone arms, respectively, and the difference was still statistically significant (P < 0.01). The proportion of patients who achieved MRD negativity in the peripheral blood was higher in the combination arm (19%) than the ibrutinib monotherapy arm (2%; P < 0.01).

While not powered for secondary outcome measures, a trend was observed towards improved PFS with IB + UTX (HR, 0.559). Median TTR for IB + UTX was 1.97 months compared to 3.8 months for single-agent IB.

Jeff Porter Sharman concluded the abstract presentation by stating that combining UTX with IB results in superior responses compared to IB monotherapy without displaying any new safety concerns. A question that remains is whether these improved responses with the combination therapy are durable. We discussed the GENUINE trial with Professor Sharman in a video interview at the 2017 ASCO Annual Meeting, watch here.

Abstract 7505

The last oral abstract on CLL was presented by Matthew Steven Davids, MD, MMSc, from the Dana-Farber Cancer Institute/Harvard Medical School, Boston, MA, USA.

Novel agents against B-Cell Receptor Kinases and BCL-2 (such as venetoclax) have demonstrated ability to improve outcomes in patients with CLL. Unfortunately, Richter’s Transformation (RT; a rare complication of CLL which occurs in approximately 5–10% of cases) to a more aggressive, fast-growing form of Large Cell Lymphoma has been observed in patients treated with novel agents, and survival of these patients is poor. Thus far, the incidence of RT in patients treated with novel agents has not undergone systematic evaluation.

During this abstract presentation, Dr Davids presented results from a retrospective study carried out on patients, diagnosed between 2011 and 2016, with pathologically-confirmed RT from 9 different academic centers.

Overall, 71 CLL patients who experienced RT while being treated with novel agents were identified. The median age of patients at CLL diagnosis was 55.3 years, ranging from 20.9 to 81.6 years. Patients had received a median number of 3 previous treatments (range, 0–12) before administration of a novel agent. Over two-thirds of patients (68%) were refractory to fludarabine and 7% of patients (n=5) had relapsed after undergoing allo-SCT. The median time from diagnosis of CLL to treatment initiation with a novel agent was 5.7 years, ranging from 0.9 to 20.5 years.

FISH analysis at treatment initiation with novel agent found that 49% (30/61) of patients harbored del(17p), 25% (15/61) harbored del(11q), and 25% (15/61) had trisomy 12. Three-quarters of patients (40/53) were found to have complex karyotype. Unmutated IGHV was present in 88% of patients (46/52); VH1-69 in 23% (10/43), VH4-39 in 9% (4/43).

When stratified by type of novel agent received, 83% of patients (n=59) were administered a BTK inhibitor, 8% of patients (n=6) were administered a PI3K inhibitor, and 8% of patients (n=6) were administered venetoclax (BCL-2 inhibitor). 

The histologies of RT were found to be DLBCL in 87% of patients, plasmablastic in 6%, Hodgkin in 4%, and other in 3%. The Ki67 proliferative index of RT were: >90 in 23% of patients, 75–90 in 25%, 50–75% in 25%, and <50 in 28%.

Median time from initiation of novel agent treatment to diagnosis of RT was 9.1 months, ranging between 0.9 and 48.2 months. Nearly two-thirds (65%) of patients developed RT within 1 year of beginning novel agent therapy.

Nineteen different initial RT treatments were reported in 56 patients and included R-EPOCH (36%), R-CHOP (20%), checkpoint blockade (9%), OFAR (7%), or a different novel agent (7%).

In this retrospective analysis, ORR was 42% (20/56) in patients who had received treatment for RT. CR rate was low with RT therapy, at 15% (7/56), and PR rate was 27% (13/56). SD or PD was reported in 58% of patients (28/56). In the 29 evaluable patients who were administered R-EPOCH/CHOP as initial RT therapy, ORR was 48% with a 21% CR rate. It was found that type of therapy for RT after novel agent does not impact ORR.

After a median follow-up of 10.6 months, median OS was poor at only 3.3 months (95% CI, 2.2­–6.0); encouragingly, of the 7 patients who achieved CR, none have died. Two patients in response went on to undergo allo-SCT: the PR patient is alive more than 3 years after allo-SCT, the CR patient is alive nearly 2 years after allo-SCT. Moreover, OS was not significantly different in patients who experience RT with different novel agents.

Matthew Davids concluded this abstract presentation by stating that, to the authors’ knowledge, this is the largest study of the development of RT on novel agent therapy in patients with CLL. It was found that patients frequently had high-risk characteristics, especially complex cytogenetics. Moreover, it was common for patients to experience RT within the first 12 months of treatment with a novel agent. Currently, there is a huge variation in treatment and, in patients who do not reach CR, prognosis remains poor. It is crucial that molecular drivers of RT are identified and standardization of treatment is undertaken. Lastly, novel therapeutic strategies to prevent and combat RT is a pressing area of unmet need.