Lenalidomide: A double-edged sword for concomitant multiple myeloma and post-essential thrombocythemia myelofibrosis.

Abstract

A 69-year old Caucasian man presented at our Institution in 2011 for a thrombocytosis discovered in the past year (platelet count 600–650 × 109/L). His medical history included habitual smoking, and a transitory ischemic attack (TIA) that occurred six years before, prompting starting therapy with acetylsalicylic acid (ASA, ie, 100 mg/daily). Physical examination was unremarkable. At referral, blood tests confirmed mild thrombocytosis (platelet count 688 × 109/L) without any other hematologic abnormality: hemoglobin (Hb) was 14.3 g/dL and white blood cells (WBC) were 8.8 × 109/L with a normal differential count. Secondary causes of thrombocytosis were excluded, serum lactate dehydrogenase (LDH) level was increased (371 U/L, reference range 135–225 U/L). Thrombocytosis is defined by a platelet count exceeding 450 × 109/L. Based on its pathogenesis, thrombocytosis can be classified into primary (clonal) and secondary (or reactive). The latter one accounts for 80% to 90% of cases, and main underlying causes include iron deficiency, infectious causes (soft-tissue, lung, and gastrointestinal infections), chronic inflammatory and autoimmune disorders, tissue damage (trauma or post-operative state), asplenia, cancer, acute blood loss. Once secondary causes are ruled out based on anamnesis, appreciation of normal iron state and negativity for inflammation markers, a myeloproliferative neoplasm (MPN) such as essential thrombocythemia (ET), primary myelofibrosis and initial phases of polycythemia vera has to be suspected.1, 2 The persistence of thrombocytosis over time and a mild elevation of LDH level are suggestive of an MPN. After excluding conditions responsible for reactive thrombocytosis in the patient, a diagnostic work up for a MPN was performed. Bone marrow (BM) biopsy documented hypercellularity (>80%) with marked proliferation of megakaryocyte cell lineage with predominance of large forms with hyper-lobulated nuclei, without fibrosis (MF-0). Cytogenetic analysis showed a normal karyotype, with negativity for BCR-ABL1 rearrangement by fluorescence in situ hybridization (FISH). No mutation in JAK2 (V617F) and MPL were documented. A diagnosis of ET was done according to 2008 WHO criteria.3 The patient was considered at high-risk based on the thrombosis score (age older than 60 years, plus a minor cerebrovascular event) and cytoreductive therapy with hydroxyurea (HU) was started along with low-dose ASA. Two months later, the treatment with HU was withdrawn because of fever and severe cutaneous toxicity, manifesting as oral aphthous stomatitis and hand and peri-malleolar skin ulcers. He refused further treatment apart from aspirin and entered strict follow-up. The patient met all criteria for a diagnosis of ET, according to the 2008 WHO criteria: However, owing to the histopathological features, particularly the increased BM cellularity, it is surmisable the correct diagnosis should be reconsidered as a prefibrotic myelofibrosis (pre-MF) by the application of the updated WHO 2016 criteria.4 In 2011, conventional risk factors for thrombosis in ET were an age greater than 60 years and a history of thrombosis. In high-risk disease, as our patient, the combination of hydroxyurea and low-dose ASA is the standard of care, recommended on the basis of controlled studies demonstrating a significant reduction in risk of thrombosis by application of cytoreduction. Further studies highlighted that JAK2 wild-type patients showed a risk of thrombosis lower than JAK2V617F mutated, leading to the development, in 2012, of a three-tiered International Prognostic Scoring of thrombosis in ET (IPSET-thrombosis); the score included JAK2 mutation status and presence of cardiovascular risk factors as additional risk variable on top of age and prior thrombosis.5 More recently the score was re-analysed leading to a refined four-tired version, which excluded cardiovascular risk factors evaluation.6 Currently, the patient would have been categorized as a high-risk both for IPSET score and its revised version, requiring cytoreduction. In any case, should the patient had been classified as pre-MF, he would receive the same treatment, since the IPSET thrombosis score was validated also in pre-MF patients.7 Hydroxyurea (HU) is the most commonly used agent for the long-term treatment of patients with MPN and it is well tolerated in the large majority of subjects. Severe side effects are reported in about 5% of patients, among which fever, pneumonitis, and cutaneous manifestations are the most frequent. In a large series of patients, HU-related fever was reported as a rare and rapid onset idiosyncratic event, typically subsiding after discontinuation and recurring after re-challenging, leading invariably to the withdrawal of the drug due to intolerance.8 According to the European Leukemia Net (ELN) Consensus Criteria,9 HU-related fever is a criterion for defining HU intolerance, identifying patients who need to stop the treatment and ultimately require a second-line therapy. A year later, in June 2012, a serum protein electrophoresis (SPEP) performed during a follow-up examination, documented the presence of a double monoclonal component (MC) IgA lambda and IgG lambda of 0.09 and 1.73 g/dL, respectively; Bence Jones protein (BJP) was absent. The patient repeated the blood and urine tests six months later. A stable IgG lambda M-protein was detected, BJP was negative, lambda free light chains was 36 mg/L and free light chain ratio (FLC) was 18; this laboratory picture was compatible with a monoclonal gammopathy of undetermined significance (MGUS). The patient continued the regular six-monthly follow-up. MGUS is defined as non-IgM serum MC less than 3 g/dL with monoclonal plasma cells less than 10% in the BM and absence of myeloma defining events (MDE) or CRAB criteria. In our patient, the BM biopsy had been performed one year earlier with no evidence of plasma cell (PC) infiltration and was not repeated. Analysing risk factors for progression in MGUS (M-protein >1.5 g/dL, non-IgG M-protein, and FLC ratio < 0.26 or > 1.65), our patient resulted at high-intermediate risk with an estimated 20-years risk of progression to multiple myeloma (MM) of more than 30%.10 In May 2016, four years later, the patient started to complain of asthenia, presenting a progressive decline in performance status; systemic symptoms and palpable splenomegaly, were absent. A careful review of blood tests revealed a slow and steady decrease in hemoglobin level over the past year, in parallel with an increase of M-protein. A new clinical reassessment was carried out. Blood tests revealed Hb of 11.0 g/dL, WBC 7.8 × 109/L, platelet count 637 × 109/L, LDH level 409 U/L. IgG lambda was 2.58 g/dL, FLC lambda and kappa were 96.7 mg/L and 3.42 mg/L, respectively; FLC ratio < 100, BJP was positive. A BM biopsy was performed, that documented trilinear hyperplasia with a cellularity of 80% (Figure 1(A)), infiltrate of lambda clonal PC accounting for 30% of total cellularity (Figure 1(B)) and grade 2 fibrosis (Figure 1(C)). FISH analysis highlighted chromosome 17 trisomy and excluded MM high-risk (HR) translocations; the search for 5q31 and 7q31 deletions, MECOM, MLL, ETV6 and RARA rearrangements by FISH were negative. Type 1 CALR mutation [52-bp deletion (p. L367fs*46)] was detected in peripheral blood with a variant allele frequency (VAF) of 57% by capillary electrophoresis (CE). A skeletal evaluation by whole-body positron emission tomography/computed tomography (PET-CT) scan excluded bone lesions. A diagnosis of concomitant post-ET myelofibrosis (PET-MF) and smoldering multiple myeloma (SMM) was done.10, 11 Mild anemia was considered as a consequence of PET-MF process, rather than to the underlying SMM, and the patient maintained a close follow-up due to the absence of other MDE.12 Cytogenetic abnormalities are found in most MM patients and FISH has a greater sensitivity than conventional cytogenetics to detect aberrations owing to the low proliferation rate of PC. Translocations involving the immunoglobulin heavy chain gene (IgH) locus and trisomies are considered primary cytogenetic alterations. Among all prognostic factors described in MM, FISH abnormalities have been found to be the most predictive of outcomes. Translocation t(4;14), t(14;16) and t(14;20) associate with poor prognosis identifying HR disease, whereas t(11;14), t(6;14) and/or trisomies, as in our patient, are considered to mark a standard-risk (SR) disease.13 In 2013, mutations in CALR (chr. 19p13.2), the gene encoding the endoplasmic reticulum-associated chaperone calreticulin, were described in about 20% of patients with JAK2/MPL unmutated ET. To date, more than 50 different mutations, all in exon 9, have been described. Two CALR mutations account for approximately 80% of all the subtypes; type-1 is a 52-bp deletion (c.1092_1143del p. L367fs*46) and type-2 is a 5-bp insertion (c.1154_1155insTTGTC p. K385fs*47), resulting in mutant proteins that lost the ER-retention motif (KDEL) at the C-terminus. All other mutations are grouped as type 1-like and type 2-like in relation to their corresponding structural similarities and effect on C-terminal.14 Based on the International Working Group-Myeloproliferative Neoplasms Research and Treatment (IWG-MRT) expert consensus (Table 1), a diagnosis of post-ET MF was performed by considering the BM histology, development of anemia and increased LDH level.11 We used the MYSEC-prognostic model, which includes advanced age, hemoglobin level below 11 g/dL, platelet count below 150 × 109/L, circulating blast cells equal to or greater than 3%, CALR-unmutated genotype and the presence of constitutional symptoms as core risk factors, to risk-stratify the patient.15 Based on this score, the patient was assigned to the low-risk category. As per anemia, which would have allocated MM into symptomatic disease implying the need for treatment, we attributed that as due to MF component, in view of the relatively low percentage of plasma cell infiltration (30%), together with the stability of M protein. As such, anemia was not considered a CRAB criterion; no other factors (CRAB-MDE) emerged for a symptomatic disease.12 In May 2018, the patient experienced intense back pain interfering with daily activities. Blood tests were stable: Hb was 11.2 g/dL, LDH 400 U/L and platelet count 538 × 109/L. Also, M-protein was 2.59 g/dL, BJP resulted positive, free lambda light chains were 95 mg/L, FLC ratio < 100. A CT scan of the spine and pelvis revealed several lytic lesions, also confirmed by a subsequent PET-CT scan. Therefore, in September 2018 he started therapy for symptomatic MM with lenalidomide, 25 mg orally daily on days 1 to 21, and dexamethasone 40 mg weekly in 28-day cycles. According to Italian reimbursement criteria, current standard induction regimens for transplant ineligible patients consist of bortezomib, melphalan and prednisone (VMP) and lenalidomide-dexamethasone (RD). When appraising the risks and benefits of treatment, we judged that alkylating agents, embedded in the VMP regimen, might be associated with excessive myelosuppression in the light of the concurrent myelofibrotic process; alkylating agents were also excluded because of their leukemogenic potential. Further, the choice of lenalidomide-based treatment was supported by studies describing the role of IMiDs in the treatment of myelofibrosis.16 The treatment was well tolerated except for a mild gastrointestinal toxicity; the patient reported rapid improvement in pain and asthenia. We observed an improvement in the Hb value in parallel with a decrease in LDH levels. After one year of treatment, SPEP documented a reduction in MC up to 0.18 g/dL, normalization of FLC and negativity of BJP. Blood tests showed resolution of anemia (Hb 13.1 g/dL), normalization of leukocytes and platelet count (4.60 × 109 and 215 × 109/L, respectively) and LDH (Table 2). We repeated BM biopsy that showed reduced cellularity (<30%) with restoration of normal hematopoiesis (Figure 1(D)), including megakaryocyte number and morphology, with virtual absence of megakaryocyte clusters, reduction in clonal PC count (<5% of all nucleated cells; Figure 1(E)) and resolution of fibrosis (grade 0; Figure 1(F)). Interestingly, a significant reduction of CALR del52 allele burden on peripheral granulocyte cell fraction by CE was documented (VAF of 16% as compared to 57% two years before). According to International Working Group (IWG) response criteria for MF17 and MM,18 the patient displayed a complete clinical remission for the former (with a partial molecular remission due to a decrease ≥50% in CALR del52 allele burden) and a very good partial response for the latter (due to ≥90% reduction in serum M-protein). Currently the patient is in good general condition, asymptomatic. One year after the start of treatment, the patient stopped steroid and lenalidomide was reduced to 10 mg/daily. Last blood tests are summarized in Table 2; MC is stable at 0.15 g/dL. In patients with MPN the risk of second primary malignancies is significantly increased as compared with the general population, in particular, cancers in the upper gastrointestinal tract and lung.19 There are several case reports and reviews describing patients with coexisting MPN and plasma-cells diseases. A monoclonal gammopathy (MGUS) can be identified in 3%–14% of MPN patients, an incidence slightly higher than expected in the general population. Note, MGUS tend to occur across all MPN entities without evidence of clonal relationship between the two diseases or any known correlation with specific genetic background.20 The pathogenetic mechanisms behind this higher risk are largely elusive but probably relate to a combination of genetic predisposition, acquired mutations and side effects of cytoreductive treatments. Moreover, an impaired immune-surveillance, intrinsic to the MPN and/or facilitated by concurrent cytoreductive treatment, may play a significant role in lymphomagenesis. Multiple myeloma has also been reported in MPN patients. It is rarely reported in patients with PV and is preferentially diagnosed in female patients without the JAK2V617F mutation.20 Most diagnoses of MM are synchronous with that of MPN or occurred in the follow-up21. Currently, no specific strategies for the management of MPN patients with a concomitant MGUS/MM have been established and defining which disease dominates the clinical picture can prove challenging due to large overlapping of characteristics. To further complicate the physician's task, a watchful wait approach can be reasonably applied in smouldering/low-risk phases of both diseases, while at the opposite timely anti-neoplastic treatment may be crucial to tackle impending clinical complications. The efficacy of IMiDs is well established in MM; although not included among standard therapeutic options, IMiDs were explored in MF setting about a decade ago: some retrospective studies in MF patients receiving these agents documented a higher response rate in the presence of mutations involving spliceosome SRSF2, U2AF1, ZRSR2 and SFRB1.22, 23 In this respect, a targeted high-throughput sequencing on PB of our patient, at the time of MF evolution, did not reveal any additional mutation. The IMiDs have demonstrated clinical activity in ameliorating anemia, thrombocytopenia and splenomegaly due to their anti-inflammatory, anti-proliferative and anti-angiogenic effects through inhibition of NF-kB mediated pro-inflammatory and apoptotic circulating cytokines, including IL-2R, IL-6, IL-10, IL-8, TGF-β and TNF-α, which are all significantly increased in MF patients.24 Some clinical trials have examined lenalidomide with or without short-term steroids in MF reporting resolution of leukoerythroblastosis and reduction in BM fibrosis and in JAK2V617F VAF.25-28 However, despite the promising preliminary data, no clear consensus for IMiDs in MF treatment has emerged. The choice of IMiDs for the treatment of concomitant MM and myelofibrosis is supported by the current case report that showed remarkable treatment efficacy in both diseases. While the efficacy of lenalidomide in MM is well known, we were particularly intrigued by the complete reversal of fibrosis and normalization of BM histology, together with normalization of blood cell counts and a significant reduction of CALR del52 allele burden. G.G.L. and E.A. analysed and interpreted the data and wrote the manuscript. I.R. collected the data and contributed to writing the paper. F.V. performed histopathological analyses. G.R. performed molecular analyses. F.M., P.G. and A.M.V. contributed to interpret the data and critically reviewed the manuscript. The authors declare no conflict of interest. not applicable