Tasquinimod in the treatment of castrate-resistant prostate cancer – current status and future prospects

Abstract

Treatment options have significantly expanded in recent years for men with metastatic castration-resistant prostate cancer (mCRPC), with the routine use of immunotherapy (sipuleucel-T) and novel hormonal agents such as enzalutamide and abiraterone acetate prior to taxane-based chemotherapy or radium-223 radiotherapy. A number of immune checkpoints limit the immune response of the host to metastatic tumor progression in prostate cancer, one of which is an immunosuppressive and pro-angiogenic cell called the myeloid-derived suppressor cell (MDSC). Tasquinimod is a small molecular oral inhibitor of S100A9, a key cell surface regulator of MDSC function, and has shown anti-angiogenic, antitumor and immune-modulatory properties in preclinical models of prostate cancer and other solid tumors. A large randomized phase II trial of tasquinimod in men with chemotherapy-naïve mCRPC demonstrated a significant prolongation in radiographic and symptomatic progression-free survival compared with placebo, which was also associated with improvements in overall survival. Tasquinimod was studied in a global phase III randomized trial in men with bone mCRPC and, while it significantly improved radiographic progression-free survival, this did not result in an overall survival benefit. However, tasquinimod is under evaluation as well as a combination therapy with other systemic agents in prostate cancer and as a single agent in other solid tumors. This review encompasses the preclinical and clinical development of tasquinimod as a therapy for men with prostate cancer.

Background

Prostate carcinoma remains the most common noncutaneous malignancy in men in the United States, with 220,000 cases diagnosed in 2014 [Siegel et al. 2014]. The estimated incidence of death from this disease is estimated at 27,540 patients annually, a number that continues to fall by 3–4% each year [Siegel et al. 2014]. Targeting the androgen receptor axis has been classically known as an effective strategy in recurrent and metastatic prostate adenocarcinoma, and novel anti-neoplastic drugs have been developed and approved to target the androgen receptor as well, including enzalutamide and abiraterone [Beer et al. 2014; de Bono et al. 2011]. However, patients eventually experience progression of disease in spite of these newer androgen receptor axis targeting therapies and this has spurred the discovery of other mechanisms of action that have antineoplastic activity. These include immunologic therapies such as sipuleucel-T, ipilimumab and Prostvac, and antiangiogenic molecules such as cabozantinib [Thoreson et al. 2014]. But despite novel hormonal agents, taxane chemotherapy and radium-223 therapy, men with metastatic castration-resistant prostate cancer (mCRPC) face a short survival of 1–3 years depending on context, illustrating the need for novel approaches to systemic therapy. One compound under active phase III investigation in prostate cancer, with a novel mechanism of action, is tasquinimod [Isaacs et al. 2006].

Tasquinimod development

Tasquinimod (Figure 1) was originally identified as a potential anticancer compound during a drug screening program for anti-inflammatory molecules and was observed to promote rather than inhibit inflammation [Isaacs, 2010]. This finding catalyzed the study of tasquinimod in humans, which was demonstrated to retain some proinflammatory activity but with reduced auto-immunity, but also retained potent antitumor activity in prostate cancer [Isaacs, 2010]. The primary antitumor mechanism of action is felt to be via engagement of the S100A9 protein, which is an immunomodulatory protein found in the tumor microenvironment and on myeloid-derived suppressor cells [Bjork et al. 2009; Hermani et al. 2005].

Figure 1.

Chemical structure of tasquinimod: 4-hydroxy-5-methoxy-N,1-dimethyl-2-oxo-N-[(4-trifluoromethyl) phenyl]-1,2-dihydroquinoline-3-carboxamide).

The S100A9 protein has been found to be a ligand for receptor for advanced glycation end products (RAGE) and to have a highly specific interaction with Toll-like receptor 4 (TLR-4) [Bjork et al. 2009]. In general, the referenced myeloid-derived suppressor cells (MDSC) have their growth and phenotype induced by chronic inflammation, which is accomplished by signaling through TLR-4 [Bunt et al. 2009]. Preclinical work demonstrated that interleukin-1β inflammation led to MDSC producing increased interleukin-10, thereby promoting a type 2 immune response, which fosters tumor progression [Becker, 2006].

In prostate cancer specifically, the calcium-modulated proteins S100A8 and S100A9 along with RAGE are upregulated in both prostatic intraepithelial neoplasia and adenocarcinoma, whereas benign prostatic tissue showed minimal to no expression [Hermani et al. 2005]. In malignant prostate tissue, when S100A9 interacts and subsequently modulates RAGE, there is activation of multiple cellular pathways including mitogen-activated protein kinases (MAP kinase) and nuclear factor κB. This consequently directly affects cell survival, invasion/metastasis and inflammatory response [Taguchi et al. 2000]. Furthermore, blockade of RAGE along with its colocalizing polypeptide amphoterin in mice with implanted tumors led to decreased cancer growth and metastases [Taguchi et al. 2000]. Based on these series of findings, it has also been hypothesized that MDSC are a prospective inflammatory checkpoint to target and employment of tasquinimod may suppress MDSC, which subsequently may kindle an antitumor response [Bunt et al. 2009]

Preclinical studies have also shown that tasquinimod modulates the anti-angiogenic molecule thrombospondin-1 in prostate tumor LNCaP cells [Olsson et al. 2010]. Following exposure to tasquinimod, thrombospondin-1 mRNA was demonstrated to be upregulated. In the same series of experiments, human prostate tumors implanted in mice were also administered tasquinimod and a multitude of effects were seen: (1) tumor growth inhibition; (2) upregulation of thrombospondin-1; (3) downregulation of hypoxia-inducible factor 1α (HIF-1α); and (4) downregulation of androgen receptor. Furthermore, evidence of antiangiogenic activity was seen, as tumor vascular endothelial growth factor (VEGF) levels were either stable or decreased upon treatment with tasquinimod [Olsson et al. 2010]. In addition, thrombospondin-1 also inhibits neovascularization via blockade of VEGF receptors on endothelial cells [Doll et al. 2001].

In addition, tasquinimod has demonstrated anti-angiogenic properties in a number of vascular assays, including VEGF-independent mechanisms. While the exact mechanisms remain unclear, there are data supporting its role in inhibiting the angiogenic switch mediated by tumor hypoxia, perhaps by regulating epigenetic effectors of angiogenesis such as histone deacetylase (HDAC) 3/4 [Isaacs et al. 2012]. In addition, MDSCs are known to express S100A9 and are pro-angiogenic. This immunosuppressive cellular checkpoint is increased in many solid tumor settings, including prostate cancer, and thus tasquinimod’s anti-angiogenic effects may in part be mediated through an indirect immunomodulatory function and inhibition of MDSC activity.

Put together, tasquinimod has a unique antineoplastic mechanism, via both direct antiproliferation and antitumor properties perhaps triggered through engagement of the S100A9 protein, as well as immune modulatory and indirect antiangiogenic effects, in prostate cancer compared with other compounds currently available in the clinic.

Preclinical data

The development of tasquinimod was launched via studies in rat prostate cancer models, spearheaded by the Isaacs laboratory at Johns Hopkins University in conjunction with Active Biotech in Sweden. The predecessor molecule to tasquinimod, named linomide (4-hydroxy-N,1dimethyl-2-oxo-N-phenyl-1,2-dihydroquinoline-3-carboxamide), had shown antineoplastic effects via antiangiogenic activity in prostate cancer xenografts, but further study with linomide was terminated due to autoimmune and proinflammatory and cardiovascular toxicity [Ichikawa et al. 1992; Noseworthy et al. 2000].

On account of these promising anticancer properties, however, 11 derivative quinolone-3-carboxamide compounds were tested to see if one could be found that retained antiangiogenic ability but without the deleterious proinflammatory stimulation [Isaacs et al. 2006]. Chemically, linomide was modified by way of substitutions in the aromatic rings. Then, each of these molecules was tested in vivo to measure the inhibition of prostate cancer growth in the Dunning R-3327 AT-1 rat model. In addition, each molecule’s potency in inhibiting angiogenesis was also was analyzed through in vivo assays in mice. The results from these two series of experiments allowed for a ranking of the 11compounds, and the one that emerged as the lead compound was ABR-215050 [4-hydroxy-5-methoxy-N,1-dimethyl-2-oxo-N-[(4-trifluoromethyl) phenyl]-1,2-dihydroquinoline-3-caboxamide]; this was later named tasquinimod.

The next critical step in the preclinical phase was to examine the inflammatory properties of ABR-215050, because the proinflammatory aspects of linomide mediated the toxicities leading to its discontinuation. ABR-215050 (tasquinimod) was assessed in a Beagle dog model validated for measuring inflammatory response and a lack of proinflammatory effect was observed [Snyder et al. 1995]. Specifically, after intravenous injections of 1 mg ABR215050/kg/day for 5 days, white blood cell count (increase of 800/ml) and erythrocyte sedimentation rate (no increase) results supported this absence of induced inflammation in this model [Isaacs et al. 2006]. This further provided support for continuing development.

ABR-215050 then underwent robust characterization of its dose–response ability to abrogate prostate cancer growth in a total of four human and rodent prostate cancer models, as well as additional assays to measure antiangiogenic proficiency. To measure the potency of inhibiting cancer growth, the total daily dose that inhibits tumor by 50% (ED₅₀ concentration) was calculated. Based on the ED₅₀ values for both the human and rodent prostate cancer models, ABR-215050 is approximately 30–60 times as efficacious as linomide [Isaacs et al. 2006].

In the four angiogenesis assays studied, results suggested at least a 50% inhibition of angiogenesis in the presence of ABR-215050, independent of effects on the immune components of the tumor microenvironment. For example, in the in vivo CAM assay in chicken eggs, 10 µM of ABR-215050 was co-administered with 100 ng of VEGF. At 48 h, the angiogenic response decreased by more than 50% [Isaacs et al. 2006]. Importantly, among the human and rodent prostate cancer models, it was observed that degree of antineoplastic effect was proportional to the antiangiogenic activity. For example, in the models examined in the Isaacs lab at Johns Hopkins, a tumor blood vessel density diminution of 44–65% corresponded to a 41–78% reduction in tumor weight.

To further investigate the antiangiogenic mechanism, tumor tissue oxygenation was also measured. CWR-22Rv1 tumor bearing mice were given either 10 mg ABR-215050/kg/day versus nothing for 1 month. Subsequently, utilizing a fiberoptic probe placed in the center of tumors, the partial pressure of oxygen (pO₂) was recorded. It was found that in treated tumors, the mean pO₂ was 5.6 mmHg compared with 15.1 mmHg in untreated tumors. Statistical analysis calculated a p value of <0.05 for this comparison.

The Isaacs lab further studied tasquinimod’s mechanism of antineoplastic activity in a hypoxic microenvironment and to test the hypothesis if tasquinimod is maximally effective in such a tumor environment [Isaacs et al. 2012]. Two cell lines, LNCaP and HUVEC (human umbilical vein endothelial cells) were subjected to growth media with a low oxygen, high carbon dioxide and low glucose mixture. Cell survival time was quantified with and without treatment with tasquinimod 1 µmol/l, a concentration that interestingly had no anticancer effect in vitro but was therapeutic in vivo. Without treatment, cell growth slowed in a statistically significant fashion when measuring doubling times, but the cells survived. With tasquinimod treatment, however, doubling time slowed further, and 52 ± 9% of LNCaP cells died in 1 week [Isaacs et al. 2012].

It has been shown that, in hypoxic conditions, gene transcription changes via global deacetylation, with increased expression of survival genes [Denko et al. 2003]. The Isaacs lab further explored this in prostate cancer cell lines (with 2% O₂ conditions) in comparing deacetylation between controls to treatment separately with tasquinimod and with trichostatin A; this latter compound is a pan-class I/II HDAC inhibitor [Haberland et al. 2009]. Results of Western blot testing showed similar diminution of histone deacetylation between trichostatin A and tasquinimod. This suggested that tasquinimod harbors HDAC inhibitory properties in the hypoxic microenvironment, thus epigenetically inhibiting an angiogenic switch in tumors during hypoxic conditioning.

To examine exactly how tasquinimod would be functioning as a HDAC inhibitor, the Isaacs lab focused on HDAC4. HDAC4 is overexpressed by 3–5 times in CRPC cells compared with benign prostate tissue [Halkidou et al. 2004]. Cell line studies have demonstrated that prostate cancer growth is inhibited by suppression of HDAC4 [Cadot et al. 2009]. Further effort showed with in vitro studies that tasquinimod binds to HDAC4, and specifically to a regulatory zinc-binding domain within the HDAC4 domain [Isaacs, 2013]. However, the clinical relevance of this mechanism in humans remains unclear.

Spurred by this robust preclinical testing of efficacy and evidence for its multiple mechanisms of action, tasquinimod (ABR-215050) moved forward to phase I trials.

Clinical trials: phase I

A phase I multicenter clinical trial of tasquinimod in Europe was conducted with patients enrolling from 2005 to 2007 [Bratt et al. 2009]. In this effort, 32 patients participated between two parts to the trial. The first part was a dose-escalation study to determine the maximum tolerated dose (24 patients) and the second was an intrapatient dose-escalation study (eight patients). Any Grade 3 or 4 toxicity was considered a dose-limiting toxicity (DLT). The concept for the intrapatient stepwise dose-escalation portion was based on observed changes in laboratory variables and toxicity without dose escalation, and to see if this more gradual dose titration approach could reduce the frequency of these changes related to inflammation and tolerability. Both nonmetastatic [rising prostate-specific antigen (PSA)] and mCRPC patients were allowed on study; key exclusion criteria included symptomatic disease, prior chemotherapy or other systemic therapy other than androgen deprivation treatment, and systemic corticosteroid use.

In the first portion of the phase I trial, 17 patients received a fixed dose of 0.5 mg orally daily and seven patients 1 mg daily (both on a continuous dosing schedule for 28 days). The 0.5 mg dose level did not experience a DLT. In the 1 mg cohort, two DLTs were observed: one patient had an asymptomatic CTC Grade 3 amylase elevation on day 23 of the tasquinimod, and the second patient had Grade 3 arrhythmia at day 10 with periods of supraventricular tachycardia. Based on this, the maximum tolerated dose (MTD) of tasquinimod was determined to be 0.5 mg orally daily without dose escalation, and 1.0 mg orally daily if within-patient dose escalation was performed (0.25 mg for 2 weeks followed by 0.5 mg for 2 weeks, followed by 1 mg daily). This latter schedule was chosen as the recommended phase II dose based on improved tolerability and the ability to continue tasquinimod at lower doses in those patients who could not tolerate the full MTD.

Safety results from the first 24 patients showed inflammation as the most common adverse event (AE) (41% of patients) with increased laboratory values of white blood cells, erythrocyte sedimentation rate and C-reactive protein, consistent with induction of a proinflammatory response. Other notable events were myalgias in 18%, fatigue in 18% and anemia in 18%. Amylase elevations were seen in 12%. Data from the second portion (intrapatient dose-escalation) revealed the most common AEs to be myalgia in 25%, hypoesthesia in 25% and pain in extremity in 25%. In this part of the study, three severe adverse events (SAEs) were recorded in two patients: one had Common Terminology Criteria (CTC) Grade 2 musculoskeletal pain and CTC Grade 3 chest pain, and a second had a cerebral infarction noted. Workup of the patient with the cerebral infarction included a computerized tomography (CT) scan negative for ischemic stroke or hemorrhage, and he reportedly recovered in 3 days. For the trial as a whole, 76% of AEs occurred by day 35 of tasquinimod treatment [Bratt et al. 2009].

Though the primary purpose of the phase I study was not efficacy, noteworthy results included 56% of patients not experiencing PSA progression by 18 weeks, and two patients had a PSA decrease of at least 50%. Furthermore, 12 of 15 patients did not develop any new bone lesions [Bratt et al. 2009]. However, given that radiographic assessments were not routinely performed in this study and the lack of a control group, conclusions about efficacy were not possible.

Clinical trials: phase II

Based on the phase I trial results, a randomized, double-blind, placebo-controlled phase II trial was conducted [Pili et al. 2011]. This was a multicenter, multinational trial across 45 centers in the United States, Canada and Sweden. A total of 201 men with mCRPC were randomized in a 2:1 fashion to either tasquinimod (134 patients) or placebo (67 patients). Utilizing the intrapatient dose-escalation model from the phase I trial by Bratt and colleagues, patients began treatment with tasquinimod taken orally starting at 0.25 mg daily and then escalation to 1 mg daily over 4 weeks. Specifically, tasquinimod was administered at 0.25 mg daily for 2 weeks, then 0.5 mg daily for 2 weeks, and then 1 mg daily.

Key inclusion criteria included histologically confirmed prostate adenocarcinoma and a Karnofsky performance score of 70–100, with castrate serum levels of testosterone (⩽50 ng/dl). All patients had to have radiographically confirmed metastatic disease. For enrollment, progression had to have occurred, as defined by: (1) rising serum prostate-specific antigen (3 consecutive measurements within 1 year at least 14 days apart); (2) progression of measurable soft-tissue metastasis; or (3) new bone lesions on bone scan within 12 weeks of screening [Pili et al. 2011]. In addition, patients had to be asymptomatic or minimally symptomatic from their prostate cancer, as indicated by a visual analog scale for pain of 0–4 on a 10 point scale and lack of chronic opiate consumption for cancer pain.

Important exclusion criteria included opiate intake, prior cytotoxic chemotherapy within 3 years, and prior anticancer therapy using biologic or vaccine treatment within 6 months. In addition, men with either a history of pancreatitis or cardiovascular disease (myocardial infarction, congestive heart failure, ventricular arrhythmias or unstable angina) were excluded. Antiandrogen use was permitted [Pili et al. 2011].

Treatment on the phase II study was divided into an initial double-blind therapy (maximum of 6 months), followed by open label medication. Asymptomatic patients in the placebo group experiencing disease progression during the first 6 months, or without progression at 6 months, were allowed to switch to open-label tasquinimod. Those on the tasquinimod arm during the double-blind portion of the trial without disease progression at 6 months were offered the chance to continue on open-label treatment until progression. The primary endpoint was the proportion of patients who were progression-free at 6 months [Pili et al. 2011].

Disease progression was defined to include at least one of the following: regular need of narcotic analgesia (either a single intravenous dose of narcotic pain medication, or more than 10 days of oral narcotic analgesic usage); palliative radiation therapy for tumor site pain; progression by Response Evaluation Criteria In Solid Tumors (RECIST) criteria; visual analog scale pain rating of more than 4 secondary to malignancy; pathologic fracture needing radiation or surgery; and spinal cord compression [Pili et al. 2011]. Two or more new bone lesions seen on bone scan could also constitute progression, as long as the finding was not consistent with tumor flare per Prostate Cancer Working Group 2 (PCWG2) criteria [Scher et al. 2008].

Phase II trial: results

The 201 patients were overall well-balanced on study, with a few clinically meaningful differences. However, the tasquinimod arm featured more patients with tumor pain (28% versus 11%), visceral metastases (24% versus 15%) and African-American patients (15% versus 3%). The primary endpoint, which was the proportion of patients progression-free at 6 months, was superior in favor of tasquinimod: 69% of patients were progression-free at 6 months compared with 37% on the placebo arm [relative risk (RR) = 0.49, 95% confidence interval (CI): 0.36–0.67; p < 0.001]. The median progression-free survival (PFS) was calculated to be 7.6 months for tasquinimod and 3.3 months for placebo [p = 0.0042, with hazard ratio (HR) = 0.57; 95% CI: 0.39–0.85] [Pili et al. 2011].

Of those patients with progression, 84% of patients had radiographic progression and 22% had symptomatic progression. Radiographic PFS (rPFS) was 8.8 months for tasquinimod and 4.4 months for placebo (HR = 0.54; 95% CI: 0.36–0.82). In terms of radiographic response by RECIST criteria, 7% of tasquinimod patients had a partial response (0% in the placebo arm) and 52% had stable disease (versus 31% on placebo). PSA kinetics were not altered on the whole compared with placebo; 4% had a PSA decline of at least 50% compared with no such declines in the comparator group.

Subgroup analyses compared outcome data among multiple subtypes, including for patients with visceral metastases, bone metastases and lymph node-only disease. These data favored tasquinimod in each category: median PFS was 6.0 versus 3.0 months (p = 0.045, HR = 0.41; 95% CI: 0.16–1.02) for those with visceral metastases; median PFS 8.8 versus 3.4 months (p = 0.019, HR = 0.56; 95% CI: 0.34–0.92) for patients with bone metastases; and median PFS 6.1 versus 3.1 months (p = 0.54, HR = 0.73; 95% CI: 0.27–2.00) for lymph node-only metastases [Pili et al. 2011]. It is important to note that these are intriguing trends in terms of general treatment effect, but phase II study was not powered to detect statistically significant differences for each subgroup. These data suggested that tasquinimod had a clinically relevant impact on PFS in men with mCRPC, particularly those with bone metastases.

Phase II safety results

The overall discontinuation rate due to toxicity was 22% for tasquinimod versus 1% on placebo. The most frequent reason for stopping therapy was muscle or joint pain. Interestingly increases in C-reactive protein were seen to have a correlation with joint pain. The majority of AEs were Grades 1–2 (89% of AEs on tasquinimod). Among Grades 3–4 AEs, the most common among tasquinimod patients were anemia (4%), asthenia (1%), deep vein thrombosis (5%) and increased lipase (7%) [Pili et al. 2011]. Cardiovascular events were rare, occurring in only 1% of men.

Multiple asymptomatic laboratory parameter deviations were observed on study. Examples included transient leukocytosis, amylase, lipase, C-reactive protein and fibrinogen, consistent with a proinflammatory effect. For most patients with these abnormalities, parameters normalized within 6 months on continued tasquinimod therapy and were not considered clinically significant [Pili et al. 2011].

Phase II trial: long-term survival

The primary endpoint from the Phase II trial of tasquinimod versus placebo was the proportion of patients progression-free at 6 months. These data significantly favored tasquinimod at 69% versus 37% on placebo. Subsequently, an analysis was undertaken and published examining whether these delays in progression were associated with long-term survival improvements in this trial, in conjunction with biomarker analyses to attempt to discover a potential driving mechanism or predictors for the better responders [Armstrong et al. 2013].

At the time of the survival analysis, 111 patients (55%) had died of the original 201 men on study. Notably 41 patients (out of 67 on placebo) crossed over to the treatment arm. The median follow up was 37 months at data calculation. An intention-to-treat (ITT) method was used to measure survival and account for crossover. Survival results demonstrated that the median survival was 33.4 months for tasquinimod compared with 30.4 months, with a HR of 0.87 (95% CI: 0.59–1.29) (Figure 2). Among the bone metastases subgroup, survival time diverged more in favor of tasquinimod at 34.2 months, contrasting with 27.1 months on placebo, with an HR of 0.73 (95% CI: 0.46–1.17). However, given that there were multiple imbalances in prognostic factors at baseline in this trial, a multivariate analysis was conducted for overall survival that analyzed the contribution of tasquinimod to survival after adjusting for key prognostic factors, such as lactate dehydrogenase (LDH), pain, PSA, PSA doubling time and hemoglobin. In this exploratory analysis, tasquinimod was found to have a PFS and overall survival advantage; the HR for PFS was 0.52 (95% CI: 0.35–0.78; p = 0.001) and for overall survival was 0.64 (95% CI: 0.42–0.97); p =0.034) [Armstrong et al. 2013]. Overall, these data were suggestive of a survival benefit, although the ITT analysis and multivariate analyses were not sufficiently powered for a definitive answer.

Figure 2.

Overall survival results from the phase II trial of tasquinimod versus placebo.

Hazard ratio = 0.87 (95% confidence interval: 0.59–1.29), p = 0.49.

A focus of the survival investigation was with correlation with biomarkers that are either known prognostic factors for castrate-refractory prostate cancer, or related to the mechanism of tasquinimod from preclinical studies. With tasquinimod, a transient increase in levels of thrombospondin-1, VEGF and RAGE were seen from the serum. A sustained increase in C-reactive protein was also observed in treated patients. Furthermore, bone alkaline phosphatase and LDH values stabilized while on tasquinimod, while these variables continued to worsen on placebo. Deeper exploration into trends juxtaposed with patient survival fascinatingly suggested that patients with a baseline level of thrombospondin-1 below median were predictive of a survival advantage with tasquinimod [Armstrong et al. 2013]. In preclinical human prostate tumor explant models, tasquinimod is known to upregulate thrombospondin-1, and thrombospondin-1 increases have been shown to abrogate of neovascularization and inhibition of tumor growth [Olsson et al. 2010; Doll et al. 2001]. Further validation of this potentially biomarker is warranted in larger studies.

Finally, in an analysis of the phase II randomized trial using an automated and quantitative method to assess disease burden in bone called the bone scan index (BSI), Armstrong and colleagues found that tasquinimod modestly delayed the worsening of BSI over time, consistent with the cytostatic effect observed preclinically, and validating objectively the radiographic PFS findings as determined both by local site and independent radiologists [Armstrong et al. 2014].

Phase III trial in mCRPC

Based on these preclinical and clinical observations, a phase III double-blind, placebo-controlled international trial of men with mCRPC and bone metastases was conducted. In this trial over 1200 men were enrolled [ClinicalTrials.gov identifier: NCT01234311] across North America, Europe, Asia and South America. In this trial, a focus on men with bone metastases was emphasized given the positive findings from the phase II trial, particularly in this subgroup. Enrollment was open to men regardless of symptom burden, and prior enzalutamide or abiraterone acetate was permitted. This trial was powered around overall survival, a key secondary endpoint, with a primary endpoint of rPFS. Given the cytostatic effects of tasquinimod, a possible best role for this agent is in combination with other active systemic agents in mCRPC, or used as a single agent in men with asymptomatic to minimally symptomatic mCRPC. Tasquinimod has an established and well tolerated safety profile suitable for chronic use, and thus is well positioned as an option for men prior to taxane chemotherapy to delay progression and symptomatic disease.

Although the final manuscript and presentation are not yet available, the phase III trial main results were reported in a press release on 16 April 2015. The manufacturer, Active Biotech, reported an HR of 1.09 for overall survival (95% CI: 0.94–1.28) and therefore tasquinimod did not prolong survival. However, tasquinimod did reduce the risk of radiographic cancer progression or death compared with placebo (HR = 0.69; 95% CI: 0.60–0.80), consistent with the phase II observations. The explanation for this unexpected negative result in this study is presently unclear, particularly until the full details of the results are available. One possibility is that, with the multiple drugs available in the clinic that already prolong survival, tasquinimod as a single agent alone may not have the potency to show a benefit on overall survival as many men will go on to receive life-prolonging therapies, diluting the possibility of observing a difference. Additionally, the impact of possible proinflammatory or cardiovascular toxicities leading to drug discontinuation is also a possibility; again, this is speculation until the full manuscript is available. Finally, it may be that while tasquinimod is active in delaying progression, this activity as a single agent is only modest, and only in combination will sufficient clinical improvements be seen.

Combination trials are underway to clarify the role of tasquinimod when sequenced with sipuleucel-T immunotherapy [ClinicalTrials.gov identifier: NCT02159950] and as maintenance therapy after initial docetaxel chemotherapy [ClinicalTrials.gov identifier: NCT01732549]; accrual is completed in the latter with results pending (Table 1). Further studies would be needed to examine the proper dosing, pharmacokinetics, and safety of combination approaches with enzalutamide, abiraterone acetate, docetaxel and radium-223. Preclinical studies have suggested at least an additive effect of tasquinimod when given with taxane chemotherapy [Dalrymple et al. 2007]. These studies have provided the impetus for the CATCH PC trial [ClinicalTrials.gov identifier: NCT01513733], which is a phase I study of cabazitaxel chemotherapy with tasquinimod. This trial is a Department of Defense Prostate Cancer Clinical Trial Consortium (DOD PCCTC) trial being conducted at Duke University and the University of Chicago. Combined, these trials should provide further clarity around the proper positioning of tasquinimod in the clinic.

Table 1.

Current tasquinimod clinical trials in prostate cancer that are either actively enrolling, or completed and awaiting final data.

NCT Number	Phase	Title	Status
NCT01234311	III	A Study of Tasquinimod in Men With Metastatic Castrate Resistant Prostate Cancer	Completed
NCT02159950	II	Sipuleucel-T With or Without Tasquinimod in Treating Patients With Metastatic Hormone-Resistant Prostate Cancer	Terminated
NCT01513733	1	The CATCH Prostate Cancer Trial: Cabazitaxel And Tasquinimod in Men With Prostate Cancer	Completed
NCT02057666	III	Study Of Tasquinimod In Asian Chemo-Naive Patients With Metastatic Castrate-Resistant Prostate Cancer	Active
NCT00560482	II	Efficacy Study of ABR-215050 to Treat Prostate Cancer	Completed
NCT01732549	II	A Proof of Concept Study of Maintenance Therapy With Tasquinimod in Patients With Metastatic Castrate-resistant Prostate Cancer Who Are Not Progressing After a 1st line Docetaxel Based Chemotherapy	Completed

Conclusion

Tasquinimod has unique mechanisms of action, possessing anti-angiogenic properties that are both VEGF-dependent and independent, as well as immune stimulating properties, possibly through inhibition of MDSC action, a key immune checkpoint in cancer. These properties are not prostate cancer specific, and have led to phase I–II trials of tasquinimod in other solid tumors (IPSEN) based on preclinical anticancer properties demonstrated in a number of tumor types [Isaacs, 2010]. Tasquinimod may also inhibit the hypoxia-induced angiogenic switch through epigenetic regulation of HDAC3/4 interactions, which may contribute to the observed delays in metastatic progression over time in clinical trials in prostate cancer. These diverse mechanisms make tasquinimod unique among prostate cancer therapies and distinct from the failed anti-VEGF therapies (e.g. bevacizumab, sunitinib) [Michaelson et al. 2014; Kelly et al. 2012] in men with mCRPC. However, only large controlled trials evaluating and powered around overall survival will be able to confirm and validate its role as a standard agents for men with mCRPC. These final results are anticipated to be presented in late 2015, with much clarity expected based on the detailed results of this large international trial.