Ibrutinib in PCNSL: The Curious Cases of Clinical Responses and Aspergillosis

Abstract

In this issue of Cancer Cell, Lionakis et al. demonstrate that the combination of temozolomide, etoposide, doxorubicin, dexamethasone, rituximab, and the Bruton tyrosine kinase (BTK) inhibitor ibrutinib induced frequent responses in patients with primary central nervous system lymphoma but was associated with significant toxicity, including pulmonary and cerebral aspergillosis infections.

Primary central nervous system lymphoma (PCNSL) is an aggressive type of diffuse large B cell lymphoma (DLBCL) confined to the central nervous system (CNS), including brain, spine, cerebrospinal fluid (CSF), and eyes. Treatment of PCNSL has evolved over the last decades. Most regimens include high-dose methotrexate (HD-MTX) in combination with other chemotherapeutics with or without radiation. Although PCNSLs usually respond to initial therapy, the overall prognosis remains inferior to other types of DLBCL outside the CNS. Up to 50% of patients relapse after an initial response to therapy, and 10%–15% suffer from primary refractory disease (Jahnke et al., 2006). Patients with primary refractory or relapsed PCNSL have a poor prognosis, with median survival of 2 months without further treatment (Reni et al., 1999). The optimal salvage regimen for recurrent/refractory PCNSL patients has not been established. Prospective trials using single-agent temozolomide or rituximab are modestly effective at providing a few months of progression-free survival (PFS).

The vast majority of PCNSLs are of the activated B cell (ABC) type non-germinal center DLBCL subtype (Camilleri-Broët et al., 2006). Outside the CNS, this ABC DLBCL is associated with worse outcome and frequent mutations in the B cell receptor (BCR) signaling pathway, including MYD88 and CD79B. The Bruton tyrosine kinase (BTK) is a central signaling node in the BCR pathway, making it an attractive therapeutic target. In fact, the BTK inhibitor ibrutinib has been approved by regulatory agencies for the treatment of several B cell malignancies, including chronic lymphocytic leukemia, mantel cell lymphoma, marginal zone lymphoma, and Waldenström’s macroglulinemia.

In this issue of Cancer Cell, Lionakis and colleagues report their clinical experience with a new chemotherapy regimen consisting of dose-adjusted temozolomide, etoposide, liposomal doxorubicin, dexamethasone, ibrutinib, and rituximab (DA-TEDDi-R) (Lionakis et al., 2017). They excluded HD-MTx from this regimen based on their preclinical data suggesting antagonism with ibrutinib. The authors explored this regimen in two different populations: patients with newly diagnosed PCNSL and patients with recurrent/refractory PCNSL. Patients were initially treated with a 14-day ‘‘window’’ of single-agent ibrutinib prior to initiation of DA-TEDDi-R. The authors reported an impressive 83% (15/18) response rate in patients who were initially treated with single-agent ibrutinib, with all responses being partial remissions. Responses were equally observed in patients who had pre-treatment steroids and in those who did not. After adding chemotherapy, the response rate was 86%, with the majority achieving complete remissions. The clinical response to DA-TEDDi-R is remarkable, especially in patients with refractory PCNSL, as six out of 11 patients responded to treatment that lasted for more than 6 months.

Unfortunately, this promising clinical activity was associated with significant toxicity, including invasive aspergillosis occurring in 39% of the patients. Ninety-four percent of participants developed a grade 4 neutropenia, 56% developed a grade 4 thrombocytopenia, and 28% (5/18) died from treatment-related toxicities. The study enrolled five patients with newly diagnosed PCNSL, of whom two (40%) died from treatment-related toxicities. The number of treatment-related deaths is considerably higher than reported for other first-line chemotherapy regimens containing HD-MTX, including HD-MTX+Ara-C (8%) (Ferreri et al., 2009), Rituximab+HD-MTX+Temo-zolomide (2%) (Rubenstein et al., 2013), and HD-MTX+Ara-C+Thiotepa+ Rituximab (<1%) (Ferreri et al., 2016), and also considerably higher than reported for the combination of high-dose chemotherapy with stem cell rescue (7%) (Soussain et al., 2008) in relapsed/refractory PCNSL.

The frequency of invasive aspergillosis in the current study (39%) is higher than the incidence that was reported in two other studies using single-agent ibrutinib in patients with PCNSL (5%–11%) (Grommes et al., 2016, ASH 58^th Annual Meeting, abstract; Choquet et al., 2016, ASH 58^th Annual Meeting, abstract). Furthermore, with the approval of ibrutinib by regulatory agencies, thousands of patients with a variety of B cell malignancies have been treated, but only a handful of cases of fungal infections have been reported. It is tempting to speculate that the concurrent use of ibrutinib and steroids in patients with PCNSL may have contributed to the increased incidence of aspergillosis. However, this is unlikely to be the case, because the two studies with lower incidence of aspergillosis also allowed concomitant use of steroids and ibrutinib (Figure 1A). The authors provided an alternate mechanistic explanation. Mice lacking the Btk gene that were exposed to Aspergillus fumigatus had a higher mortality rate compared to mice with wild-type Btk, suggesting that a functional BTK may play a role in the innate immune control of Aspergillus infection (Figure 1B). However, because only 27% of Btk^−/− mice died of Aspergillus infection, other BTK-independent mechanisms must exist to protect against this fungal infection. Furthermore, it is unclear whether these mice would have had higher incidence of Aspergillus infection if they were treated with steroids. Regardless of the etiology, one may consider pre-emptive prophylaxis strategy to prevent fungal infection in patients with PCNSL receiving ibrutinib. However, fungal prophylaxis may interfere with ibru-tinib elimination in the liver by inhibiting CYP3A metabolism, which may increase the risk of adverse events. Therefore, additional studies are needed to improve the safety of this regimen before DA-TEDDi-R can be adapted into clinical practice. Accordingly, established HD-MTX-based regimens should remain the standard therapy for newly diagnosed PCNSL until further data about safety, tolerability, and clinical outcome is available for ibrutinib-based regimens.

Figure 1. Therapeutic Targeting of BTK in PCNSL.

(A) Activity of single-agent ibrutinib in PCNSL. (B) Ibrutinib inhibition of BTK in tumor cells induces favorable clinical responses, while inhibition of BTK in monocytes may lead to an increased incidence of invasive aspergillosis.

Lionakis and colleagues were able to examine CD79B and MYD88 genetic alterations only from four patients. Based on a meta-analysis of DNA sequencing studies in PCNSL, the expected incidence of CD79B and MYD88 L265P mutations is 56%. Given the high response rate (83%) achieved with single-agent ibrutinib that was reported by Lionakis et al., responses must have been achieved in cases lacking these mutations. In these cases, it is possible that other genetic alterations may have contributed to BCR pathway activation. However, it is unfortunate that a more comprehensive genome sequencing was not performed in all cases. BCR signaling may have been activated by self-antigens regardless of activating mutations. It is unclear whether clinical responses achieved in cases lacking MYD88 and CD79B mutations may have been induced by the use of steroids rather than ibrutinib.

The study reported by Lionakis et al. is an important step forward toward finding more effective treatment strategies for patients with PCNSL. However, more data will be needed on the safety of ibrutinib-based regimens before ibrutinib can be widely incorporated in the treatment of PCNSL.