Predictive Factors Found for Treatment With FCO in CLL

A mutation on the NOTCH1 gene was shown to be an independent predictive factor for the reduced efficacy of ofatumumab (Arzerra), a human monoclonal CD20 antibody, in the treatment of patients with relapsed/refractory chronic lymphocytic leukemia (CLL), according to study results presented at the 2017 International Workshop on Chronic Lymphocytic Leukemia. The impact of additional mutations, including TP53 and SF3B1, were also noted.

The phase III COMPLEMENT 2 trial investigated a cohort of patients with previously treated, relapsed/refractory CLL (NCT00824265). A total of 325 patients were randomized in a 1:1 ratio to receive fludarabine and cyclophosphamide (FC) alone or in combination with ofatumumab (FCO). The patients were also assessed with a custom DNA sequencing panel looking at TP53, SF3B1, and NOTCH1 mutations.

The investigators found 221 mutations in the 3 genes. There were variations in the types of TP53 mutations, but the SF3B1 mutations were exclusively missense and the NOTCH1 mutations were mostly nonsense. “When we look into the distribution of the different mutations among patients, we can see that for TP53, 20 of 61 affected patients had a second, third, or, in total, up to 6 different TP53 mutations, while a second or third mutation in SF3B1 was quite rare,” said Eugen Tausch, MD, an assistant physician in the Department of Dermatology and a research assistant at the Institute of Immunology of the University Hospital of Ulm in Germany, who presented the findings.

Ultimately, there was an incidence of 18.8% of TP53 mutations, 19.7% for SF3B1 mutations, 16.3% for NOTCH1 mutations, and a notably low incidence of 4.6% for 17p deletions. “There was still an association of 17p deletion and TP53 mutation, with a P value of 0.01,” Tausch commented. “We also found an association of NOTCH1 mutations and gain of chromosome 12q.” There were no associations between these 3 mutations and IGHV mutation status, gender, age, CLL Binet stage, ECOG performance status, systemic B symptoms, or white blood count, he added.

However, there were differences across the study between patients with wild-type or mutated genes. There was a lower complete response (CR) rate in both treatment arms combined in TP53-mutated patients compared with wild-type patients (11.7% vs 19.9%, respectively), but the difference was not found to be significant. There were also lower CR rates in SF3B1-mutated patients (12.5% vs 19.9%) and NOTCH1-mutated patients (14.3% vs 19.2%) compared with wild-type patients; these differences also were not statistically significant.

For the overall response rate (ORR) in both treatment arms, there were lower ORRs in mutated patients compared with wild-type patients for both TP53 (67.8% vs 84.4%; P <.01) and SF3B1 (71.9% vs 83.7%; P = .05) mutations, which were both statistically significant. Alternatively, NOTCH1-mutated patients showed a higher ORR than wild-type patients (85.7% vs 80.5%), although this was not deemed statistically significant.

“TP53, in this trial, is an independent adverse prognostic factor for survival, and I think it’s important to consider that most of these mutated patients were not 17p deleted,” Tausch commented.

Data regarding survival indicate that TP53 mutations are associated with a shorter progression-free survival (PFS) in both the FCO and FC treatment arms. This reduction, comparing TP53-mutant and wild-type patients, was statistically significant, with a hazard ratio (HR) of 1.93 (P <.001). Similarly, there was a significant reduction in overall survival (OS) between TP53-mutant and wild-type patients in both arms (HR, 2.11; P <.001).

SF3B1 mutations also had an impact on survival in both treatment arms, but the results were not statistically significant. Comparing SF3B1-mutant and wild-type patients, there was an HR of 1.37 (P = .07) for PFS and 1.30 for OS (P = .23). With NOTCH1, the results showed that the mutation is a predictive factor for reduced efficacy of ofatumumab in both PFS and OS. In the FCO treatment arm, NOTCH1-mutant patients had a shorter PFS compared with NOTCH1 wild-type patients, but there was no difference in PFS in the FC treatment arm.

When comparing the 2 arms, there was a statistically significant difference in NOTCH1 wild-type patients (HR, 0.64; P = .01), but there was no significant difference among NOTCH1-mutant patients (HR, 0.86; P = .67). “If you have a NOTCH1-mutated patient, he will not benefit or is not very likely to benefit from the addition of ofatumumab to FC therapy,” Tausch said.

The investigators also assessed CD20 surface levels in the patients. However, there was no difference between the NOTCH1 wild-type and mutant patients in the CD20 surface levels, as was the case for both TP53 and SF3B1.

In a multivariate analysis, the investigators looked for independent adverse prognostic factors. The IGHV unmutated gene has been previously reported as a prognostic factor, and this was echoed in PFS in the COMPLEMENT 2 trial (HR, 2.17; 95% CI, 1.61-3.03; P <.01). IGHV unmutated also impacted OS (HR, 1.85; 95% CI, 1.22-2.78; P = .01).

Additionally, a deletion on 17p was significant as an adverse prognostic factor for PFS (HR, 4.47; 95% CI, 2.51-7.92; P <.01) and OS (HR, 5.02; 95% CI, 2.69-9.36; P <.01). TP53 mutations were similarly significant as prognostic factors for PFS (HR, 1.80; 95% CI, 1.29-2.50; P <.01) and OS (HR, 1.68; 95% CI, 1.12-2.52; P = .04). Treatment including ofatumumab was also a prognostic factor for PFS (HR, 0.66; 95% CI, 0.52-0.85; P <.01), but not for OS (HR, 0.85; 95% CI, 0.61-1.17; P = .41).

Notably, the multivariate analysis did not show that NOTCH1 mutations and SF3B1 mutations were independent adverse prognostic factors for PFS or OS.

Tausch E, Estenfelder S, Dolnik A, et al. NOTCH1 mutation as a predictive factor in the context of fludarabine and cyclophosphamide (FC) with or without ofatumumab treatment. In: Proceedings from the 2017 International Workshop on Chronic Lymphocytic Leukemia; May 12-15, 2017; New York, NY.