Multikinase inhibitors for treating high-risk myelodysplastic syndromes: can this be brought into clinical practice?

Keywords::

• azanucleosides failure
• investigational therapies
• multikinase inhibitors
• myelodysplastic syndromes

The management of higher risk myelodysplastic syndromes (MDSs) has undergone a dramatic change over the last decade with the approval of the two azanucleosides azacitidine and decitabine. Both azanucleosides are generally well tolerated and have clear disease-modifying activity, but response rates are relatively low, and even patients who initially respond eventually develop disease progression. For these patients, no approved agents are currently available, leading to poor outcomes with a median survival of approximately 6 months [1].

Tyrosine kinase inhibitors (TKIs) have shown efficacy in a variety of both solid and hematologic malignancies. Perhaps the best example of therapeutic success with these agents has been in chronic myelogenous leukemia, where the targeted inhibition of BCR–ABL with imatinib and the second-generation TKIs has altered the landscape of disease management and resulted in improved long-term survival. Although many TKIs were developed as targeted therapies directed at specific enzymes, these agents also exert effects through modulation of ‘off-target’ pathways, which may be exploited for therapeutic benefit. Preclinically, erlotinib, dasatinib, sorafenib and lapatinib have all been found to induce cell cycle arrest and apoptosis in acute myeloid leukemia (AML) cell lines and/or primary bone marrow samples from AML and MDS patients. Several of these TKIs were subsequently evaluated in clinical trials for patients with higher risk MDSs with variable results.

The oral EGFR inhibitor erlotinib is currently approved for patients with nonsmall cell lung cancer harboring an EGFR gene mutation. Of particular note, several case reports describe patients who achieved complete remission (CR) from AML when given erlotinib for a concurrent lung malignancy. Further preclinical studies showed that erlotinib was capable of inducing differentiation and cell cycle arrest in EGFR-negative myeloblasts of patients with MDSs and AML, possibly through inhibitory effects on Janus kinase 2, Src family kinases and/or mammalian target of rapamycin [2]. Two clinical trials evaluating the activity of erlotinib in MDS have been reported to date. In a single-institution study, 35 MDS patients previously treated with at least one azanucleoside were enrolled [3], of which 26 had higher risk disease. The combined overall response rate was 14.7%, including 3 with marrow CRs (mCRs), 2 with hematologic improvement (HI) and 11 with stable disease. Another Phase I/II clinical trial conducted by the Groupe Francophone des Myélodysplasies specifically targeted higher risk MDS patients who had failed azanucleosides [4]. Only 3 of 22 evaluable patients had a response, including 1 mCR and 2 HI, and 6 had stable disease. In both studies, no grade 3 or 4 hematologic toxicity was seen. Common nonhematologic toxicities were consistent with those previously reported in lung cancer, including diarrhea, rash, sepsis, fatigue and anorexia. Newer preclinical data demonstrate synergism between erlotinib and azacitidine in AML and MDS myeloblasts, resulting in increased degradation of MCL-1 and BCL2L10 and upregulation of PUMA accompanied by increased intracellular accumulation of azacitidine [5]. BCL2L10 upregulation mediates azacitidine resistance and its downregulation may restore azacitidine sensitivity [6]. Combined treatment with azacitidine and erlotinib was studied in an early phase solid tumor trial; further studies are planned for patients with MDSs.

Other TKIs that have shown limited activity in clinical trials for patients with MDS include dasatinib, sorafenib and axitinib. In addition to its ability to inhibit BCR–ABL, dasatinib has activity against other tyrosine kinases including Src family kinases. Among the Src family kinases, Lyn kinase in particular is constitutively active in myeloblasts and megakaryocyte progenitors in MDS, and inhibition of Lyn kinase suppresses leukemic cell growth in vitro [7]. The activity of dasatinib as monotherapy in MDS was evaluated in a Phase II trial of 18 patients who failed azanucleosides, among whom 3 responded: 1 patient with AML had >50% reduction in blast percentage and 2 MDS patients achieved mCR without HI [8]. Another Phase II trial of dasatinib in patients with a variety of hematologic malignancies reported no responses in 6 patients with MDS or chronic myelomonocytic leukemia (CMML) [9]. Sorafenib 400 mg twice daily as monotherapy is associated with significant toxicity including fatigue, constipation and diarrhea, and is fairly toxic to elderly patients with MDSs [10,11]. Sorafenib combined with chemotherapy yielded a low response rate in unselected MDS and AML patients with its activity largely limited to AML patients with FLT3 gene mutations [12]. When combined with azacitidine, however, encouraging CR rates with acceptable toxicity was seen in patients with AML, most of whom had FLT3 mutations [13]. There is a small subset of MDS patients who harbor FLT3 gene mutations for whom further study of sorafenib combinations may be warranted. With the idea of inhibiting bone marrow angiogenesis through modulation of VEGF activity, axitinib was studied in AML–MDS, but no responses were seen [14]. Bone marrow expression of VEGFR-1 and VEGFR-2 was limited in these patients, which may partially explain the lack of clinical activity despite reductions in soluble VEGFR-2 plasma levels and elevations in plasma VEGF and placental growth factor levels.

The multikinase inhibitor furthest in development is rigosertib (formerly ON 01910.Na), a styryl sulfone derivative that inhibits both phosphatidylinositide 3-kinases and polo-like kinases. This drug selectively promotes G2/M arrest in malignant cells and synergizes with cytotoxic agents in preclinical models [15]. Four Phase I/II trials have been conducted in MDS and AML patients, and full results of three of these studies are reported in the literature. Results from the fourth trial have not been reported individually but have been included in combined analyses. Seetharam et al. conducted a Phase I/II trial of 13 red blood cell-dependent, higher risk MDS patients previously treated with hypomethylating agents, which resulted in 4 mCRs, 1 partial remission and 8 with stable disease [16]. Correlative studies after cycle 1 demonstrated decreased phospho-Akt2 in CD34⁺ cells by 12–19% in three patients with either mCR or stable disease, compared to only a 2% decrease or 19% increase in two patients with progressive disease. In an early phase trial conducted at the NIH, 12 higher risk MDS and 2 AML patients were treated with rigosertib, among whom 3 had at least a 50% reduction in bone marrow blasts and 3 patients achieved HI [17]. In another Phase I/II study of patients with MDS or AML, a total of 21 patients with higher risk MDS had previously failed azanucleoside and all AML patients previously had MDS [18]. The maximum tolerated dose was 1375 mg/m² administered as a continuous infusion for 72 h. Of 18 evaluable patients 4 patients had mCR, 2 had HI and 2 patients had >50% bone marrow blast reduction from baseline but not to <5%. Rigosertib was well-tolerated and showed little hematologic toxicity in these trials, but clinically significant urinary events including hematuria and cystitis were reported which required treatment interruption, hydration and sodium bicarbonate in some patients. Other Grade 3 or 4 nonhematologic toxicities reported include fatigue, nausea and diarrhea.

Bone marrow blast response to rigosertib was associated with improved survival in the cumulative trial experience [18,19]. The reported median survival in all three trials was approximately 9–10 months, which compares favorably with retrospective studies of unselected patients who have failed azanucelosides. However, patients who are fit enough for investigational therapy are reported to have better outcomes, with a median survival of 13.2 months [1]. With overall survival as the primary end point, a pivotal randomized Phase III trial of rigosertib versus best supportive care in patients with RAEB-1, RAEB-2, RAEB-t or CMML after failure of azanucleosides is nearing completion with results eagerly awaited.

Based on its favorable bioavailability, an oral formulation of rigosertib is also under development but may be better suited for patients with lower risk disease. In a Phase I study, dose-limiting toxicity was reached at 700 mg twice daily as a result of dysuria and shortness of breath [20]. Bioavailability of the recommended Phase II dose (560 mg twice daily) was 35% and administration with food significantly reduced the rate and extent of absorption. Decreases in blast count were seen in five out of eight patients with RAEB-1 or RAEB-2, including mCR in two patients, both of whom had low/intermediate-1 disease. Five lower risk patients experienced HI and a Phase II clinical trial in lower risk patients is currently accruing.

In summary, clinical trials of multikinase inhibitors as monotherapy in patients with higher risk MDS have been met with limited success, perhaps with the exception of rigosertib, which is currently the subject of a randomized Phase III trial. A definitive survival benefit has yet to be demonstrated for any investigational agent in this patient population. Continued efforts are needed to elucidate the mechanism of action of the multikinase inhibitors in responding MDS patients to foster individualized selection of those most likely to respond. As we learn more about novel gene mutations in critical pathways in MDS, perhaps targeting those patients with TKI pathway mutations may be the strategy in upcoming studies.

Financial & competing interests disclosure

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending or royalties.

No writing assistance was utilized in the production of this manuscript.