Impact of Vorinostat on Pruritus Signaling Pathways - Merck Study

Overview

Mycosis Fungoides (MF) is a rare malignancy in the United States. It is the most common form of cutaneous T-cell lymphoma (CTCL). Sézary syndrome (SS) is the most severe and leukemic form of CTCL. Pruritus, or itch, is defined as an unpleasant sensation that elicits the desire to scratch. Severe itch is a manifestation of all forms of MF, especially those with patch/plaque and folliculotropic variants, as well as in Sezary patients. While severe itch causes great suffering for patients, the pathogenesis of itch in MF and Sezary syndrome is complex and not well understood. It is thought that various chemical mediators are produced by the malignant cells to cause itch. Vorinostat, an FDA approved therapy for the treatment of MF, has also been reported to relieve pruritis. The goal of the study is to evaluate how vorinostat affects different chemicals in the skin that have been known to cause itch. This is a single center, non-randomized study designed to obtain and test blood and skin tissue samples take at various time-points over 6 months in patients who are prescribed vorinostat per standard of care treatment. Samples from pruritic and non-pruritic skin and blood of MF and Sezary patients will be evaluated for the presence of chemicals thought to be important in the cause of itch in these diseases. This evaluation will include immunohistochemistry, RT-PCR, and ELISA assays. The results from this study may help define how vorinostat decreases itch in patients with MF and Sezary Syndrome.

Mycosis Fungoides (MF) is a rare malignancy in the United States. It is the most common form of cutaneous T-cell lymphoma (CTCL). Sézary syndrome (SS) is the most severe and leukemic form of CTCL. Severe pruritus is a manifestation of the SS and also affects patients with MF. Severe pruritus is a manifestation of SS and also a common symptom among patients with MF. Pruritus can significantly impact on quality of life. In one national survey of the impact of MF on patients' health-related quality of life (HRQOL), the majority of survey respondents (88%) reported being bothered by pruritus. The pathogenesis of pruritus is complex and not well understood. It affects a majority of patients with mycosis fungoides (MF) and is most severe in Sézary syndrome (SS), the leukemic form of cutaneous T-cell lymphoma. Chemical or physical stimuli that perturb, without damaging, only the epidermis (the outer most layer of the skin), elicit the symptom of pruritus. The perception of itch is abolished if the epidermis is removed. This observation and the data on PAR2 expression on nerve fibers indicate that pruritus in inflammatory skin disorders including MF/SS is likely to have a neuroepidermal origin or be mediated by molecules that percolate to the epidermis from inflammatory cells in the dermis. In this regard, MF/SS are characterized histologically by infiltration of the epidermis by neoplastic T cells (epidermotropism), and in some of these cases, the neoplastic cells form collections of 2-3 cells around epidermal Langerhans cells (Pautrier microabscess). Neoplastic cells of advanced MF/SS typically are polarized to secrete TH2 cytokines. It has been hypothesized that the accumulation of neoplastic cells in skin lesions promotes a shift from a TH1 predominant cytokine profile in early MF to a TH2 predominant cytokine profile in advanced disease 4. The shift to TH2 production coincides with impaired immune responses, eosinophilia, and high levels of IgE in the blood. Severe pruritus is more frequent in patients with advanced MF/SS suggesting that tumor burden or cytokines elaborated by activated tumor cells may be correlated to intensity of pruritus. While it is not clear which cells contributed to the TH2 cytokine shift, studies of T-cells isolated from skin and blood of affected patients showed activation of the signaling molecule, STAT3. In CTCL lines, constitutively activated STAT3 mediates IL-5 production (principal regulator of eosinophilia) and IL-13 (favors antibody responses) 5. Moreover, IL-31 also seems to signal through STAT3 in one lung model. Current treatments of pruritus are aimed towards symptomatic relief although these do not address the mechanism of pruritus. If neoplastic cells are producing IL-31 or other pruritogenic factor as we propose, then any treatment that eliminates these cells will also be effective for pruritus. To date, effectively improving pruritus and the disease manifestations is not always possible especially in advanced MF or SS. Vorinostat is a FDA approved histone deacetylase inhibitor for the treatment of the cutaneous manifestations of patients with cutaneous T-cell lymphoma (MF/SS) who have progressive, persistent, or recurrent disease on or following 2 systemic therapies. Vorinostat induces differentiation, growth arrest, and/or apoptosis of malignant cells both in vitro and in vivo and has shown clinical responses in approximately 30% of patients with advanced MF/SS 7. Overall, 32% of patients with severe pruritus had relief of this symptom, and it was observed that the relief of pruritus often presaged the clinical response, especially in SS. This suggests to us that the gene modulating effects of vorinostat might be directly inhibiting expression of IL-31 or other cytokines that mediate pruritus in MF/SS. IL-31 appears to be implicated in pruritus of atopic dermatitis. 8 Recently, studies have demonstrated that itch can be mediated by a novel cysteine protease and that proteinase-activated receptor-2 plays an important role in itch signaling pathways. 9, 10 In another recent study, substance P has been implicated in itch pathophysiology. In one small case series of patients with CTCL, the efficacy of aprepitant was evaluated11. Aprepitant is a substance P antagonist that blocks the neurokinin 1 (NK1) receptor. Substance P has a role in the physiology of unmyelinated C-fibers in the skin that convey both pain and pruritus and has the ability to modulate mast cell functioning. Data from skin biopsies have shown that successful treatment with vorinostat is associated with decreased dermal and epidermal lymphocytes, a shift from nuclear to cytoplasmic p-STAT-3 staining, and decreased CD31-positive dermal vessels. This was not confirmed in another study. Vorinostat has been postulated to indirectly inactivate STAT proteins that drive cellular proliferation and TH2 cytokine expression. Researchers observed that in skin biopsies of patients enrolled in the vorinostat stage IIB trial, nuclear accumulation of STAT1 and high levels of nuclear pSTAT3 in malignant T cells correlated with a lack of clinical response. To date, there has been no evaluation of STAT3 proteins and TH2 cytokines in pruritic versus non-pruritic skin lesions of CTCL. Signal transducers and activators of transcription (STATS) are important to active transcription of cellular proteins including cytokines. Constitutive activation of STATs has been shown to be a feature of MF/SS. Study Rationale: Since pruritus is a common manifestation of the disease that significantly impacts on patient's quality of life and that oral vorinostat provides clinical relief from pruritus in treated patients, understanding the mechanisms of action of vorinostat on cytokines implicated in pruritus will shed light on its therapeutic effect. In particular, because we believe that a reduction in pSTAT3 activity as evidenced by fewer and/or less strongly stained tumor cells may be taking place in patients responding to therapy and reporting decrease pruritus. Thus, understanding the extent of STAT3 protein aberration and cytokine expression (IL-31) in pruritic skin lesions could significantly contribute to our understanding of the mechanism of pruritus in MF/SS and to how vorinostat relieves this serious symptom. Objectives/Hypothesis It is hypothesized that skin tissue expression of pSTAT3, IL-31, and cathepsin S will correlate with response to vorinostat, a recently FDA approved therapy for MF and SS, decreasing in expression with relief of pruritus. Methods: Dosing and Frequency of Drug Participants who are cared for at Boston Medical Center will first be assessed by physicians of the CTCL multi-specialty clinic if vorinostat, administered per standard of care, is an appropriate therapy for their CTCL. The decision to invite patients to participate in this study is (1) separate from the above described clinical decision to utilize vorinostat, and (2) will be offered subsequent to the clinical decision to utilize vorinostat. There is no placebo group for this study. Up to ten participants are desired for this study. Three biopsies will be taken: (1) at baseline; (2) after 4 weeks of 300 mg daily vorinostat, and (3) if tolerated, after 4 more weeks of 400 mg daily vorinostat. At each time point, up to 2 biopsies will be taken depending on if 1 or both skin types are present (itchy vs non-itchy) at that time point and whether or not itchy or non-itchy skin shows a lesion. For each time point and each skin type, pSTAT3 staining will be measured as strong (2+), weak (1+), none (0). At each time point, patients will report the extent of their pruritus on a visual analogue score (VAS) from 0-100 mm (0, no pruritus; 100, worst imaginable pruritus). Meaningful change in pruritus will be a change in VAS score of 30 mm or more from baseline to the third time point (a change measure that will also be computed for all endpoints; the baseline to time 3 measure being of primary clinical interest). Sample Size/Accrual Rate: The total enrollment planned is 10 patients at a rate of 1-2 patients per month. With 10 subjects, for our primary hypothesis (correlation of staining intensity with self-reported degree of pruritis) we can detect a Spearman rank correlation coefficient as small as 0.75 for a one-sided test with 80% power (a directional hypothesis that is reasonable) or one as small as 0.80 for a two-sided hypothesis. In each case, these coefficients are of clear clinical relevance and are plausible given the nature of the measures. For the hypothesis of equal changes between treated and control areas on the same individuals, we propose using paired data analyses such as paired t-tests and Wilcoxon rank sum tests. For a sample size of 10 and 80% power, a paired t-test can detect an effect size of 0.996 or greater for a two-sided test and an effect size of 0.853 or greater for a one-sided test. These effect sizes are large but are within reason to expect for a clinically effective treatment.