Thiotepa intrathecal injections for myelomatous central nervous system involvement

Abstract

The worldwide incidence of multiple myeloma (MM) has been rising continuously, representing an overall increase of 126% from 1990 to 2016 in almost all regions, regardless of socio-demographic status and availability of sensitive diagnostic techniques. Though remarkable progress has been made, the disease is still currently incurable. Nearly 13% of patients will develop extramedullary disease due to the natural accumulation of mutations, evolutionary pressure or the development of sanctuaries where tumour cells are protected from therapeutic drugs. In particular, central nervous system (CNS) involvement is a critical prognostic indicator, with overall survival of less than 6 months. Because of its rarity, there is no consensus on the optimal treatment. Radiotherapy is the subject of controversy, in terms of both efficacy and toxicity. In addition, most available treatments, including cytotoxic agents, monoclonal antibodies, immunomodulators and proteasome inhibitor agents, have a very-low distribution in the cerebrospinal axis, blocking an effective, curative treatment. Several studies have tested intrathecal injections of methotrexate and cytarabine, commonly used in lymphoma CNS involvement, but with limited success, largely due to the poor efficacy of these agents against plasma cell disorders. Thiotepa, a well-known alkylating agent, has been used intravenously for decades, especially for the conditioning of MM autologous stem cell transplantation (ASCT). Additionally, it has also been injected intrathecally for the treatment of leptomeningeal metastases, without any serious adverse effects, notably in the case of breast cancer. Here we have conducted a retrospective, single-centre study to evaluate the safety and efficacy of intrathecal infusion of thiotepa in 13 consecutively diagnosed cases of CNSinvasive MM (CNS-MM). This study was conducted in full compliance with the principles of the Declaration of Helsinki with oral consent of the patients. Between October 2017 and March 2020, 13 consecutive patients with CNS-MM (eight harbouring cranial nerve palsies, two altered mental status, one numb chin syndrome, one radiculoneuritis and one paresthesia) were identified at Saint Louis Hospital in Paris (Table I). CNS involvement was defined either by the presence of atypical dystrophic plasma cells in the cerebrospinal fluid (CSF) (confirmed by flow cytometry if necessary) and/or clinic-radiological manifestations. The median age of those with CNS involvement was 62 years [range 52–72 years]: three at MM diagnosis, one at relapse and nine in refractory MM (most of the time after one ASCT plus one salvage therapy). Aggressive initial disease was observed in all cases with one third of revised international staging system (R-ISS) 3 and two thirds of R-ISS 2 [three harbouring t(4;14), two del17p/TP53 mutation, two amp1q21/del1p32, one t(9;14) and one t(14;20)]. Magnetic resonance imaging or computed tomography documentations were available for 12 of the 13 patients and revealed two (15%) with intraparenchymal plasmacytomas, three (23%) with leptomeningeal and nine (69%) with osteodural infiltrations. Treatment strategies, which varied according to patient characteristics and drug availability, are shown in Table I. All patients received intrathecal injections of thiotepa and methylprednisolone in combination with systemic therapy. Thiotepa was administered intrathecally (IT) at a total dose of 10 mg associated with 40 mg of methylprednisolone weekly until a CNS response or disease progression. A median number of three injections [range 1–8] was performed for each patient. An ASCT was performed front-line for two (15%) patients with CNS-MM at diagnosis of MM. For eight (62%) patients, radiotherapy was performed for focal lesions that were observed on imaging. The CNS-MM criteria response was defined as the cytological clearance of plasma cells from the CSF, or a complete radiological response, or a partial radiological response associated with normalization of CSF protein level and/or pleocytosis and/or clinical manifestation. The systemic response was described in accordance with the criteria of the International Myeloma Working Group. A CNS response was observed in 85% of the study population: neurological symptoms regressed in eleven (85%) patients, radiological and biological responses were observed in eight (62%) cases. In five (38%) cases, CSF was cleared at the cytological level. Among the eleven responders, 80% experienced CNS relapse in a context of systemic progression. After relapse or progression of the CNS-MM disease, four of the 13 (31%) patients continued to receive second-line therapy for CNSMM and one of 13 (8%) a third-line therapy. None of the patients underwent allogeneic stem-cell transplantation. At the time of evaluation, seven (54%) patients were alive. Finally, post CNS-MM survival was calculated with a median follow-up period of 20 months [range 2–141 months]. The median post CNS-MM overall survival, using the Kaplancorrespondence