Ruxolitinib for PV and ET
A phase II trial is also evaluating ruxolitinib for patients with hydroxyurea-refractory or intolerant PV and ET (NCT00726232). Additional inclusion criteria included hematocrit (Hct) >45% or dependence on phlebotomies (for patients with PV) and platelet count >650 χ 109/L (for patients with ET). Results were presented recently in an abstract form (Verstovsek et al. 2010c). A total of 73 patients were enrolled (PV = 34; ET = 39). Best clinically active dose was determined to be 10 mg twice daily for PV and 25 mg twice daily for ET.
In PV, therapy with ruxolitinib led to a hemat-ocrit <45% in the absence of phlebotomies in 97% of patients, and all continued to maintain response at last follow-up. There was also improvement (>50%) in splenomegaly (59% of those evaluable), normalization of leukocytosis (>15 χ 109)L) and thrombocytosis (>650 χ 109/L) in 63% and 69%, respectively. Complete remission (CR) rate was 62%, and 38% had a partial response (PR), for an overall response rate (ORR) of 100%. Patients also reported improvement in pruritus, night sweats, and bone pain. Side effects included anemia (all grades, 74%; grade 3-4, 0%), thrombocytopenia (all grades, 29%; grade 3-4, 6%), leukopenia (grade 1-2, only 15%) and weight gain (grade 1-2, only 15%).
For patients with ET, therapy with ruxolitinib normalized platelet counts to < upper limit of normal after a median of 0.5 months in 49% of patients. Among four patients with palpable splenomegaly, all experienced >50% reduction or normalization in spleen size. ORR was 62% (CR: 41%; PR 21%). More common side effects were anemia (grade 1-2, only 74%) and weight gain (grade 1-2, only 23%). Grade 3 adverse events were leukopenia (N = 2), peripheral neuropathy (N = 1) and GI disturbances (N = 1).
A decrease in the JAK2 V617F allele burden of at least 20% was obtained in 42% of PV patients and 56% of ET patients. However, similar to MF, clinical responses were unrelated to the presence of the JAK2 V617F mutation and/or reduction in JAK2 V617F mutated allele burden.
Currently, there is a randomized, open-label phase III trial for patients with PV underway. The RESPONSE trial (study of efficacy and safety in polycythemia vera subjects who are resistant to or intolerant of hydroxyurea: JAK Inhibitor INC424 (INCB018424) tablets versus best available care) (NCT01243944) compares ruxolitinib (initial dose 10 mg twice daily) to best available therapy in patients with hydroxyurea-refractory or hydroxyurea-intolerant PV. Primary endpoint is the proportion of patients who achieve a response (absence of phlebotomy requirement and >35% reduction from baseline in spleen volume as determined by MRI/CT) at 32 weeks of therapy. Secondary endpoints include the percentage of patients who achieve CR at 32 weeks and the percentage of all randomized patients who both achieve response at 32 weeks and maintain that response for >48 weeks from the time response was initially documented. The study has recently started, with an estimated enrolment of 300 patients. Results are expected in 2013.
TG101348
TG101348 is an orally available, selective JAK2 inhibitor. TG101348 has selective activity against JAK2 (IC50 = 3 nM) when compared to JAK1 (IC50 = 105 nM) and JAK3 (IC50 = 996 nM) (Geron et al. 2008; Wernig et al. 2008) . In pre-clinical studies, TG101348 inhibited JAK2 activity and phosphorylation of downstream targets, induced apoptosis of JAK2 V617F-positive cells and had activity in a murine model of PV (Geron et al. 2008; Wernig et al. 2008) .
Results of a phase I/II trial with TG101348 were recently published (Pardanani et al. 2011). In this study, 59 patients with high- or intermediate-risk MF (PMF or post-PV/ET) received treatment with TG101348 in successive dose escalation cohorts ranging from 30 to 800 mg once daily, in a classic 3 + 3 design, with expansion at MTD. JAK2 V617F mutation was positive in 86% of cases, splenomegaly >10 cm could be found in 83% of patients, and 36% were transfusion dependent. The MTD was 680 mg once daily, and DLTs included asymptomatic grade 3-4 hyperamylasemia and hyperlipasemia. Most frequent non-hematological side effects included nausea (all grades, 69.5%; grade 3-4, 3.4%), vomiting (all grades, 57.6%; grade 3-4, 3.4%), diarrhea (all grades, 64.4%; grade 3-4, 10.2%) and abdominal pain (grade 1-2, only 10.2%). Myelosuppressive complications included new-onset transfusion-dependent anemia (grade 3-4, 35.1%) and thrombocytopenia (grade 3-4, 23.7%). Among patients treated at the MTD cohort, incidence of grade 3-4 anemia and throm-bocytopenia was 54.2% and 27%, respectively. Similar to other compounds, responses consisted of improvement in spleen size, systemic symptoms, and decreased PB cell counts. After six cycles of therapy, a CI in splenomegaly by IWG-MRT criteria was observed in 39% of patients, and 45% of those in the MTD cohort. There was significant improvement in systemic symptoms after 2-6 cycles of therapy: early satiety (56% complete resolution), fatigue (63% improvement), night sweats (64% complete resolution), cough (67% complete resolution), and pruritus (50% complete resolution). Normalization of WBC count and platelet count was observed in 56% and 90% of patients who presented with leukocytosis and thrombocytosis. No improvement in transfusion-dependent anemia was observed. Median JAK2 V617F allele burden at beginning of therapy was 20% (range 3-100%), with 45% of patients having an allele burden greater than 20% (median 60%; range 23-100%). There was a decrease in JAK2 V617F allele burden after six cycles, with a median value of 17% (p=0.04). Results were more dramatic for the subgroup with high JAK2 V617F allele burden, who had a median allele burden of 31% (p=0.002) after six cycles of therapy. Currently, patients enrolled in the phase I/II trial have been transferred to an extension study (NCT00724334) to evaluate long-term effects of TG101348. Future clinical trials with this drug will evaluate different dose schedules in order to decrease side effects and maintain the efficacy, followed by a phase III study for patients with MF.
CYT387
The compound CYT387 (YM Biosciences Inc., Mississauga, Canada) is a novel dual JAK1/JAK2 inhibitor in clinical trials for patients with MF. In vitro CYT387 inhibits JAK1, JAK2, and TYK2 with IC50 values of 11, 18, and 17 nM, respectively (Tyner et al. 2010), while having no activity against JAK3 (IC50 = 155 nM). CYT387 inhibited cell lines dependent on JAK2, such as EPOR-JAK2 V617F transduced Ba/F3 cells (IC50 = 500 nM), Ba/F3 cells transduced with MPLW515L (IC50=200 nM) and endogenous erythroid colonies from patients with JAK2 V617F-positive PV (Pardanani et al. 2009; Tyner et al. 2010). Inhibition of proliferation was associated with apoptosis and decreased phosphorylation of JAK2, ERK1/2, and STAT5, consistent with on target activity (Tyner et al. 2010). In an animal model of PV-like MPN, treatment with CYT387 normalized leukocytosis, erythrocytosis, and spleen size (Tyner et al. 2010). Normalization of pro-inflammatory cytokines was also observed.
Initial results of a phase I/II clinical trial with CYT387 (NCT00935987) have been presented in abstract form (Pardanani et al. 2010). Sixty patients were recruited; 80% had a spleen size >10 cm, and 55% were transfusion dependent. Twenty-three percent of patients had failed a previous JAK2 inhibitor (either TG101348 or ruxolitinib). MTD was determined to be 300 mg/day, and DLT included grade 3 headache and grade 3 hyperlipasemia at 400 mg/day. Currently, the study is being expanded at the MTD, aiming for a total of 140 patients recruited. Improvement in pruritus was documented in 92% of patients, alongside improvements in night sweats, bone pain, and fever. A reduction in splenomegaly >50% to qualify as a CI by IWG-MRT criteria occurred in 47% of patients. More strikingly was the response observed in anemia. Fifty percent of evaluable patients had an anemia CI by IWG-MRT criteria. Among 19 evaluable patients who received the MTD of 300 mg/day, 56% had a response. The response rate was 69% in 16 transfusion-dependent patients treated at the MTD. Grade 3-4 myelosuppressive complications included anemia (7%), thrombocytopenia (26%), and neutropenia (5%). Other common side effects were nausea (grade 1, only 15%), diarrhea (grade 1, only 12%), increased transaminases (all grades, 23%; grade 3, 3%) and increased bilirubin (grade 1, 16%). The beneficial effect of CYT387 in Hb level in patients with MF has not been observed with other JAK2 inhibitors in clinical trials. Further maturation of data and results of this study are awaited in the near future.
Other JAK2 Inhibitors in Clinical Development
CEP-701 (lestaurtinib) is an indolocarbazole alkaloid that has activity as a FLT3 and JAK2 (IC50 = 1 nM) inhibitor (Hexner et al. 2008). One single centre phase II trial treated 22 patients with JAK2 V617F-positive MF with CEP-701, and responses were seen in six patients only (27%), consisting of reduction in splenomegaly (N = 3), reduction in splenomegaly with improvement in cytopenias (N = 2) and transfusion independency (N= 1) (Santos et al. 2010). Main toxicities were cytopenias and gastrointestinal disturbances. Another phase I/II clinical trial (NCT00668421) is exploring the possibility that different preparation of CEP-701 might have increased efficacy in treating patients with MF. However, preliminary results have demonstrated at best modest efficacy (Hexner et al. 2009), and final results are awaited. Pharmacological issues may hamper the efficacy of CEP-701 by limiting its ability to inhibit target kinases. In a clinical trial of patients with relapsed FLT3-mutated AML who were randomized between salvage chemotherapy and salvage chemotherapy plus CEP-701, there was a strong correlation between achieving FLT3 inhibition and CR rate (39% [FLT3 inhibition] versus 9% [no FLT3 inhibition]; p = 0.004) (Levis et al. 2011). Lack of FLT3 inhibition was associated with variable levels of CEP-701, due to binding to plasma proteins such as alpha1-glycoprotein acid.
The JAK2 inhibitor SB1518 has selective activity against JAK2 (IC50 = 19 nM), and inhibits proliferation of JAK2 V617F-positive cells in association with decreased JAK2 and STAT5 phosphorylation. Two phase I clinical trials (NCT00719836 and NCT00745550) are evaluating SB1518 for therapy of MF, and preliminary results have been reported for both (Seymour et al. 2010; Verstovsek et al. 2010a). MTD was 500 mg/day in both trials, but the dose of 400 mg/ day was chosen for phase II evaluation since there is no increase in plasma drug levels above this dosage. In one study, there was significant reduction in spleen size and systemic symptoms (Verstovsek et al. 2010a). Toxicities included diarrhea (81%; grade 3, 6%), nausea (grade 1-2, only 41%) and vomiting (grade 1-2, 22%). One advantage of SB1518 over other JAK2 inhibitors is that it appears not to be myelosuppressive, and the drug can be safely used in patients with platelet count <100 χ 109/L (Seymour et al. 2010; Verstovsek et al. 2010a) . Phase 3 study of this agent in MF is planned.
The selective JAK2 inhibitor AZD1480 was shown to inhibit JAK2 activity and phosphorylation of STAT5 in Ba/F3 cells transduced with the chimeric protein TEL-JAK2 (Hedvat et al. 2009). No activity against TEL-JAK1, TEL-JAK3, and TEL-TYK2 was seen, confirming the selectivity of AZD1480 against JAK2 (Hedvat et al. 2009). AZD1480 reduced tumor growth in xenograft models of solid tumor cell lines, and the compound is currently being evaluated in a phase I/II clinical trial in patients with PMF and post-PV/ ET MF (NCT00910728).
Most JAK2 inhibitors in clinical development affect both mutated and wild-type JAK2. That would be expected since the mutation is located in the pseudokinase domain, and these drugs compete with adenosine triphosphate (ATP) for the ATP binding site on the tyrosine kinase domain. Recently, results of pre-clinical studies with the JAK2 inhibitor LY2784544 were presented (Ma et al. 2010). In vitro, LY2784544 has selective activity against the mutated JAK2 V617F. LY2784544 inhibited JAK2 V617F signaling at a concentration 41-fold lower than JAK2 wild type (IC50 = 55 nM [V617F] versus 2.260 nM [wild type]) (Ma et al. 2010). LY2784544 preferentially inhibited proliferation on JAK2 V617F-positive cells as compared with wildtype JAK2 cells (IC50 = 113 nM versus 1,356 nM). In a mouse model of MPN, LY2784544 decreased the burden of mutated JAK2 V617F cells but had no effect on unmutated erythroid cells. Currently, a phase I/II trial (NCT01134120) is underway to confirm the selectivity of LY2784544 for JAK2 V617F and determine whether this would reflect in the side effect profile of the drug.
Histone Deacetylase Inhibitors
Rationale
Epigenetics are biochemical modifications of chromatin that regulate gene expression without affecting DNA sequencing (Herman and Baylin 2003). Main epigenetic mechanisms are DNA methylation and post-transcriptional histone modifications (e.g., acetylation, methylation). Epigenetic abnormalities in cancer cells can lead to silencing of tumor suppressor genes (Herman and Baylin 2003) . Several post-transcriptional histone modifications can occur which modulate gene expression. Histone acetylation of lysine residues is typically associated with increased gene expression (Marks et al. 2001). Histone acetylation is modulated through acetylases (HATs, histone acetyltransferases) and deacety-lases (HDACs, histone deacetylases). Besides histones, HDACs can deacetylate several other target proteins, including transcription factors (e.g., p53, STAT1, STAT3), signaling mediators, steroid receptors, chaperone proteins (heat shock protein 90 [Hsp90]) and DNA repair enzymes (Ku70) (Xu et al. 2007). Acetylation of the chap-erone protein Hsp90 prevents its association with client proteins, such as BCR-ABL and JAK2 (Bali et al. 2005; Wang et al. 2009) . Inhibition of Hsp90 leads to client protein degradation by the proteasome (Bali et al. 2005) .
A series of epigenetic abnormalities has been described in Ph-negative MPNs. We have already mentioned the effect of JAK2 V617F on histone H3Y41 phosphorylation status and its implication for gene expression regulation (Dawson et al. 2009). Wang et al. described increased activity of several histone deacetylases, particularly those of class I (HDACs 1, 2, 8) and class III, in MF cells (Wang et al. 2008) . In another study, increased expression of genes for HDACs 6, 9, and 11 were found in samples from patients with MPNs compared to controls (Skov et al. 2010). More recently, mutations of genes involved in epigenetic regulation of gene expression have been described in patients with Ph-negative MPNs, including EZH2 (Ernst et al. 2010), TET2 (Ko et al. 2010) and DNMT3A (Abdel-Wahab et al. 2011).
HDAC inhibitors (HDACIs) are compounds which inhibit activity of HDACs and can lead to increased histone acetylation and gene expression. In pre-clinical studies, several HDAC inhibitors have demonstrated activity against JAK2 V617F-positive cells. Some of HDACIs in clinical trials for MPNs include givinostat (formerly known as ITF2357; Italfarmaco, Milan, Italy) and panobinostat (formerly known as LBH589; Novartis, Basel, Switzerland). Givinostat is a synthetic class I HDACI (Guerini et al. 2008; Rambaldi et al. 2010). In vitro, givinostat preferentially inhibited proliferation of JAK2 V617F cells compared to JAK2 wild-type cells (Guerini et al. 2008). This was associated with decreased intracellular levels of total and phosphorylated JAK2 V617F and phosphorylated STAT3 and STAT5. There was no change in JAK2 V617F mRNA, indicating that down-modulation of JAK2 V617F occurred most probably at the protein level (Guerini et al. 2008). The HDACI panobinostat induced JAK2 V617F degradation by the proteasome and was synergic with the JAK2 inhibitor TG101209 in inducing apoptosis of BM mononuclear CD 34+ cell from patients with MF (Wang et al. 2009) . Wang et al. also demonstrated that panobinostat inhibited JAK2 V617F phosphorylation and phosphorylation of downstream messengers, including STAT3, STAT5, Akt, and ERK1/2 (Wang et al. 2009).
Current Clinical Trials
A pilot study with givinostat in MF was recently published (Rambaldi et al. 2010). Twenty-nine patients with JAK2 V617F-positive MPNs (PV = 12, ET = 1, MF = 16) received treatment with givinostat 50 mg twice daily. Responses were seen in 54% of PV/ET patients (CR = 1, PR=6) according to European LeukemiaNet criteria (Barosi et al. 2009). Responses in MF were more modest, only three of 16 patients had a major response. Side effects included diarrhea (62%, mostly grade 1-2), anemia (21%, mostly grade 1-2), thrombocytopenia (10%), fatigue (17%), and QTc elongation (17%) (Rambaldi et al. 2010). There is a phase II clinical trial with givinostat ongoing for patients with PV who did not respond to hydroxyurea (NCT00928707). Patients will be randomized between two doses of givinostat (50 mg once or twice daily) to be used in combination with the maximum tolerated dose of hydroxyurea already in use before study beginning. Endpoints include the response rate at 12 weeks of therapy, safety, and molecular responses at 24 weeks of treatment.
Panobinostat was evaluated in a phase I trial for patients with hematological malignancies, including 13 patients with MF (DeAngelo et al. 2009). Most patients received panobinostat thrice weekly, and the MTD was 60 mg/dose. Grade 3-4 side effects included thrombocytopenia (33%), fatigue (28%, DLT), neutropenia (28%), and anemia (12%). Among the 13 patients with MF, four had CI by IWG-MRT criteria, including one patient who became transfusion independent. These results were followed by two studies evaluating panobinostat solely in patients with MF (Mascarenhas et al. 2009; DeAngelo et al. 2010). Mascarenhas et al. reported on a phase I study in 15 MF patients (NCT01298934); the drug was given at doses ranging from 20 to 30 mg thrice weekly, and MTD had not been determined. Thrombocytopenia was the DLT. Responses were seen in four patients, all CI with reduction in splenomegaly (Mascarenhas et al. 2009). In another trial presented last year (NCT00931762), 31 patients with MF were treated with panobinostat (initial dose: 60 mg thrice weekly) (DeAngelo et al. 2010) . Treatment with panobinostat led to depletion of phosphorylated JAK2 V617F, STAT3, STAT5, Akt, and ERK1/2. Interestingly, there was a decrease in the allele burden ranging from 10% to 90%. Clinical response data is still not mature but will be presented in the near future.
Mammalian Target of Rapamycin Inhibitors
Rationale
The PI3K-Akt-mTOR pathway is abnormally activated in several types of cancer (Panwalkar et al. 2004) . In erythroid cells, activation of the PI3K pathway occurs in response to EPO stimulation and leads to resistance to apoptosis and increased proliferation by downregulation of the cyclin-dependent kinase inhibitor p27 (Bouscary et al. 2003) . PI3K can be activated directly by JAK2, and this leads to downstream activation of the serine/threonine kinase Akt and subsequently mTOR (Al-Shami and Naccache 1999). Activated mTOR is a serine/threonine kinase which regulates cell growth, proliferation, and metabolism (Panwalkar et al. 2004). Disruption of the PI3K-Akt-mTOR pathway can be demonstrated in JAK2 V617F-positive MPNs, which show increased phosphorylation and activity of Akt (James et al. 2005; Grimwade et al. 2009).
Current Clinical Trials
Rapamycin (sirolimus) is the first mTOR inhibitor to be discovered; it is currently used mostly as an immunosuppressant in solid organ transplantation. More recently, new mTOR inhibitors were developed, such as everolimus (RAD001; Novartis), which is a 40-O-(2-hydroxyethyl) derivative of sirolimus. Everolimus was tested in 30 patients with MF in a phase I/II clinical trial (Vannucchi et al. 2010). The MTD was determined to be 10 mg/daily. Most common side effects included mouth ulcers, hypertriglicerydemia, and anemia. Fifty-two percent of evaluable patients reported disappearance of systemic symptoms, and pruritus improved in 74% of cases. According to IWG-MRT criteria, the CI rate was 23%, with reduction in splenomegaly. Correlative studies demonstrated a reduction in levels of cyclin D1 (an mTOR target gene) but no changes in JAK2 V617F allele burden (Vannucchi et al. 2010).
Immunomodulatory Drugs
Rationale
Immunomodulatory drugs (IMiDs) are medications which modulate immune responses and have diverse anti-cytokine and anti-angiogenic activities (Corral et al. 1999). In Ph-negative MPNs, particularly MF, there is a systemic deregulation of the immunologic milieu, with increased levels of pro-inflammatory and fibrogenic cytok-ines such as IL-2R, IL-8, IL-12, IL-15, transforming growth factor-beta1 (TGF-ß1), bone morphogenetic protein 1 (BMP1), tumor necrosis factor-a (TNF-a), and vascular endothelial growth factor (VEGF).Pro-inflammatory cytokines are related to development of systemic symptoms and are associated with worse survival in MF .There are three available IMiDs: thalidomide, lenalidomide, and pomalidomide. Thalidomide was the first IMiD to be evaluated in MF (Barosi et al. 2001). Response rates, by IWG-MRT criteria, are in the range of 28-32%, either alone or in combination with prednisone (Jabbour et al. 2009; Thapaliya et al. 2011). Lenalidomide is a much more potent thalidomide derivative. Three clinical trials with lenalidomide have been published in MF (one single agent; two in combination with predni-sone).Response rate is 23-30%, with improvements seen in anemia (19-30%) and splenomegaly (10-42%) (Quintas-Cardama et al. 2009; Mesa et al. 2010).
Current Clinical Trials
Pomalidomide (formerly CC-4047; Celgene) is a thalidomide analog which has 10,000-fold greater anti-TNF-a activity (Galustian et al. 2009) . The first published study of pomalidomide evaluated 84 patients with MF who were randomized among four arms: pomalidomide 2 mg daily plus placebo (N = 22), pomalidomide 2 mg daily plus prednisone (N = 19), pomalidomide 0.5 mg daily plus prednisone (N = 22) and prednisone alone (N = 21).Observed benefit was limited to improvement in Hb, and response rates were 23% (pomalidomide 2 mg+placebo), 16% (pomalidomide 2 mg+prednisone), 36% (pomalidomide 0.5 mg+prednisone) and 19% (prednisone alone). Thrombocytopenia (11%) and anemia (10%) were the most common side effects, but overall, the drug was very well tolerated. In a phase I/II trial, the MTD of pomalidomide was determined to be 3.0 mg/day, and DLT was myelosuppression (Mesa et al. 2009). In accordance with the results of the randomized study, better response rates were observed in patients who received low-dose pomalidomide (63% CI in anemia).
Currently, there are four ongoing studies with pomalidomide for MF. Begna et al. reported in abstract form the results of a phase II study of low-dose pomalidomide alone (0.5 mg/day) in 58 patients with MF (NCT00669578) (Begna et al. 2010) . The main eligibility criterion was transfusion dependence or Hb < 10 g/dL. The rate of CI in anemia was 17% by IWG-MRT criteria. An increase in platelet count in thrombocytopenic patients was observed in 58% of cases. No patient had an improvement in splenomegaly. Interestingly, the JAK2 V617F mutation was a positive predictive marker of response (RR 24% [positive] versus 0% [negative]; p=0.03). Responders developed baso-philia in the first month of therapy (RR 38% [present] versus 6% [absent]; p=0.02) and did not have marked splenomegaly (38% versus 11%; p=0.05). Two other phase II studies of similar design are currently being conducted (NCT00946270 and NCT00949364), but no results are available at the time of this writing. A phase III randomized study of pomalidomide versus placebo in patients with MF who are transfusion dependent (NCT01178281) is also underway. Patients will be randomized to either pomalidomide (0.5 mg/day) or placebo; primary endpoint is rate of transfusion independence after 6 months of therapy.
Anti-fibrotic Drugs
Rationale
As mentioned, in MF, there is an increased production of cytokines associated with BM fibrosis and osteosclerosis, such as TGF-ß and BMP-1 (Castro-Malaspina et al. 1981; Kimura et al. 1989; Martyre et al. 1994; Rameshwar et al. 1994; Ciurea et al. 2007). In mouse models of thrombopoietin-induced MF, transplantation of TGF- ß-null murine hematopoietic cells retrovirally transduced with murine TPO does not elicit BM fibrosis into recipient mice, despite eliciting other features of the disease including leukocytosis and thrombocyto-sis (Chagraoui et al. 2002). This suggests that TGF-ß has a central role in the pathogenesis of fibrotic BM seen in this disorder.
Current Clinical Trials
GC1008 (Genzyme, Cambridge, MA) is a monoclonal antibody (IgG4) that targets and binds TGF-ß proteins. Results of a phase I study in patients with metastatic, advanced melanoma and renal cell cancer were presented (Morris et al. 2008) . The MTD was 15 mg/kg/dose. No DLT was observed. Side effects included rash, fatigue, headache, epistaxis, and diarrhea (Morris et al. 2008). Currently, a phase I trial (NCT01291784) in patients with MF is being conducted. Patients will receive a starting dose of 1 mg/kg/dose every 4 weeks for six applications.
Bisphosphonates
Rationale
In MF, increased bone formation with osteoscle-rosis can be observed in BM biopsies (Thiele and Kvasnicka 2005), and patients frequently complain of bone pain (Mesa et al. 2007). Histomorphometric studies in MF have revealed an uncoupling between bone formation and reabsorption, with a resultant increase in osteo-blasts’ activity and a decreased number of osteoclasts (Schmidt et al. 2007). The dysplastic megakaryocytes of MF probably play a role in the abnormal bone metabolism observed, as mega-karyocytes can increase osteoblasts activity and inhibit osteoclasts (Kacena et al. 2006). In two mouse models of MF, the GATA-1 low and the TPO-high mice, there is an increased number of megakaryocytes, accompanied by an increased bone mass and osteosclerosis (Yan et al. 1996; Vannucchi et al. 2002; Kacena et al. 2004). Increased levels of TGF-ß and osteoprotegerin are probably related to the increased in bone formation seen in MF (Chagraoui et al. 2002, 2003; Kakumitsu et al. 2005).
Bisphosphonates (BPPs) are organic analogs of the pyrophosphate molecule, with a carbon atom substituting the central oxygen atom. BPPs inhibit bone absorption by osteoclasts, re-establishing the equilibrium between bone re-absorption and bone formation (Rodan and Fleisch 1996) . BPPs are used for treating several non-malignant and malignant conditions, such as osteoporosis, Paget’s disease of the bone, solid tumors bone metastasis, and multiple myeloma. Even though there is increased bone formation in MF, BPPs indirectly diminish bone formation by decreasing bone absorption and bone turnover (Chavassieux et al. 1997) . In this regard, BPPs have been successfully used for treatment of predominantly osteoblastic (with increased bone formation) metastatic lesions of prostate cancer. Thus, BPPs could potentially have a role in MF by reverting the increased bone turnover seen in this disorder (Diamond et al. 2002) .
Current Clinical Trials
At the present time, activity of BPPs in patients with MF is limited to case reports. Sivera et al. reported on a patient with MF who was treated with oral etidronate (6 mg/kg/day on alternate months) due to severe bone pain (Sivera et al. 1994). After 3 months of therapy, there was complete disappearance of pain and fever, and this was associated with a significant improvement in Hb and WBC. In another report, therapy with clodronate (30 mg/kg/day) led to transfusion independence and reversal of BM fibrosis in a single patient with MF (Froom et al. 2002) . In other reports, there was efficacy of BPPs for treating bone pain associated with MF (Perkins et al. 2004; Assous et al. 2005) . Currently, there is one clinical trial with BPPs ongoing in MF (NCT00287261). In this international, multicenter phase II trial, patients with MF requiring therapy will receive zoledronic acid (4 mg/dose) infusions every 3 weeks for a total of 12 infusions. Primary endpoint is the effect of zoledronic acid on Hb level and spleen size. Other endpoints include measures of safety and toxicity in this patient population, effect on transfusion needs, LDH, WBC count, platelets, constitutional symptoms, performance status, and BM histology. Trial has been completed, but no results have been presented so far.
Conclusions
In recent years, the outlook for patients with chronic Ph-negative MPNs has changed with new discoveries on molecular biology of these diseases and the subsequent development of new compounds directed against those molecular defects. With standardization of diagnostic and response criteria, the reporting of clinical trials will be more uniform and will facilitate analysis of results obtained with different agents. JAK2 inhibitors and other compounds, such as pomalidomide, are on the verge of making the transition from clinical trials to routine clinical use. There is still much room for improvement though. We need to better understand how these drugs work, and through which mechanism(s) they are producing benefits for individual patients. We need biomarkers to determine which patients will respond to a specific agent. And, similarly to conventional chemotherapy used for treating solid tumors and other hematological malignancies, we need to study combination therapy, to see if there is any improvement in the response rate when the medications are combined. With coordinated efforts and appropriate design of clinical trials, we can hope to overcome these challenges and improve outcomes for patients with these diseases, not just in controlling disease signs and symptoms but potentially in eliminating it.
