The HDAC are a large family of enzymes that catalyze the removal of acetyl groups from substrate proteins and thereby regulate their function and activity (16
). Several lines of evidence suggest a role for histone acetylation in tumorigenesis. First, overexpression of HDACor mutation or deletion of their counterparts, the histone acetylases has been observed in a variety of human cancers. In addition, several proteins that have been implicated in tumorigenesis, including p53, BCL6, Hsp90, Stat3, and E2Fs, are regulated by acetylation. Finally, several oncogenic proteins and fusion proteins interact with HDACto mediate inappropriate repression of target genes, and restoration of aberrant gene expression has been shown to be critical in the control of cancer cell proliferation.
Preclinical and early clinical data support the potential utility of a variety of HDAC inhibitors for treatment of human cancers. Suberoylanilide hydroxamic acid was well tolerated and both hematologic (22% overall response rate) and solid tumor responses (12% OR rate) were observed (18
). This compound (also known as Zolinza or vorinostat, Merck & Co.) has recently been approved for use in cutaneous T-cell lymphoma based on strong phase II findings (19
). Similarly, depsipeptide has shown early promise in the treatment of refractory chronic and acute leukemias (20
). Frequent DLT include myelosuppression, nausea, vomiting, dehydration, diarrhea, anorexia, fatigue, asthenia, and electrolyte disturbances. Histone acetylation in PBMC and tumor biopsies was detected following treatment and pharmacokinetic data indicate that serum concentrations of HDAC inhibitors can be achieved well in excess of the concentrations that are necessary for cytodifferentiation in vitro
MS-275 is unique among HDAC inhibitors in clinical development, in that it inhibits class I HDAC(HDAC 1 and 3) more than class II (HDAC 4), displays a unique in vitro
cytotoxicity profile in the NCI COMPARE algorithm, inhibits tumor cell growth in nude mice comparable with or better than conventional cytotoxic agents such as 5-fluorouracil, and has relatively limited clinical toxicity, thus making it an attractive agent for development in humans (2
This phase I study was done to evaluate the toxicity and pharmacokinetic profiles of MS-275 on three different schedules: one dose every other week, twice-weekly dosing for 3 weeks followed by 1 week of rest, and weekly dosing for 3 weeks followed by 1 week of rest. No DLT were noted on the every other week dosing schedules at doses ranging from 2 to 6 mg/m2. On the twice-weekly dosing schedule, patients were treated at 2 mg/m2, and grade 3 hypophosphatemia and hyponatremia and grade 2 transaminitis (aspartate aminotransferase) were noted. Although these toxicities were not dose limiting, they resulted in dose delays and omissions in the first cycle, which were felt to preclude further dose escalation on this schedule.
For patients treated on the weekly administration schedule, 4 mg/m2 was felt to be tolerable. Patients treated at 5 mg/m2 experienced grade 3 asthenia and hypophosphatemia (despite adequate oral phosphorus supplementation). Overall, hypophosphatemia occurred frequently and may be a class effect, as it has been seen with several other HDACinhibitors, although the mechanism remains undefined. In this study, grade 2 and 3 hypophosphatemia was noted in all patients treated on the twice-weekly schedule, whereas it was less frequent on the other schedules.
No hematologic DLT were observed in this study. However, the rules for dose reduction in the case of hematologic toxicities were relatively conservative, requiring dose reduction after any cycle with grade 3 or 4 toxicity. There did not appear to be any relationship between prior therapy received and specific toxicities or DLT noted. Patients who were more heavily pretreated did not experience toxicities or DLT at a higher rate than did patients who had less treatment before study entry.
No significant cardiac events, electrocardiogram changes, or MUGA abnormalities were noted in this study. Byrd et al. noted few cardiac effects with depsipeptide given to patients with refractory acute and chronic lymphoid leukemias (22
), whereas another study of depsipeptide in patients with refractory solid tumors identified frequent, reversible ST-T wave changes, and one patient treated was reported to have grade 4 atrial fibrillation (23
). In the NCI study with MS-275, dose-limiting supraventricular tachycardia was noted on the daily × 28 schedule, but no symptomatic cardiac events were noted in patients who received alternate-week MS-275 (14
Pharmacokinetic studies revealed that MS-275 was rapidly absorbed under fasting conditions reaching a tmax within 60 min of dosing for all schedules. Thereafter, concentrations declined biphasically, with a fast elimination phase within 4 h and a slow terminal elimination phase with distribution into tissues. All patients receiving 4 mg/m2 had a detectible plasma drug level 168 h after dosing. Patients who received 6 mg/m2 (n = 4) attained a Cmax with a larger coefficient of variation between patients (285%) than in either 2 or 4 mg/m2 dose levels. Similar to the 4 mg/m2 dose level, biphasic elimination was noted, with faster clearance within 4 h, and two patients having detectible levels 168 h after dosing. Terminal half-life values could not be determined in the majority of patients, as the perceivable linear portion of the curve was <2 half-lives. Thus, the calculated t1/2 values, AUC, and clearance were estimated and can only been seen as general reference values. Similar to the study of Ryan et al., the terminal t1/2 values for MS-275 were estimated to be 60 to 150 h. This was noted regardless of the dose and schedule when the perceivable part of the curve could be followed for a relatively long time (168 h).
On the every other week administration schedule, the Cmax and tmax samples were likely underestimated with the first time-point collection at 1 h following administration as evidenced by the shorter time to maximal concentrations noted in the cohorts where earlier sampling was done (0.25 and 0.5 h for the weekly and twice-weekly cohorts). In these latter cohorts, Cmax values were observed within 0.5 h in 13 of 17 patients (76%).
Due to toxicity in patients, and subsequent holding of doses on the twice-weekly and weekly administration schedules, some pharmacokinetic variables could not be completed. This deviation from the planned schedule may have led to an underestimate of the exposure of MS-275 following repeated administration. The individual terminal half-life values could not be determined appropriately in all patients, because the time interval used for half-life estimation was too long. Thus, the mean terminal half-life values and the mean AUCand apparent clearance were estimated. For two of six patients in whom individual terminal half-lives could be determined, the values were slightly shorter compared with the values for patients in whom t1/2 was estimated (63.4 and 31.5 h, respectively, compared with 52–150 h for the population estimates). In the 5 mg/m2 weekly cohort, all patients held the second and/or third dose due to toxicity, and some patients required dose reduction. As a result, deviations prevented accurate estimation of the exposure of MS-275, and descriptive statistics were not calculated for this cohort.
Accumulation of MS-275 in plasma following weekly and twice-weekly administration was evaluated to a limited extent due to deviation form the dosing schedules required for toxicity. Based on the limited data, plasma concentrations of MS-275 appeared to increase continuously up to the last dose in cycle 1, indicating that no steady state was achieved in each cycle of the twice-weekly treatment. Conversely, there was no indication of drug accumulation on the weekly schedule.
Finally, it is difficult to evaluate dose-proportionality due to variability in the Cmax and AUC0–48 h and the small patient numbers investigated over a range of doses and schedules. The Cmax and the early part of the AUC were poorly defined, because the first sampling point was taken after the tmax on the every other week schedule. On this schedule, there appear to be dose-dependent but overproportional increases in Cmax and AUC0–48 h. However, when the twice-weekly and weekly schedules were evaluated, the Cmax and AUC0–48 h values increased almost dose-proportionally after single administration of 2 to 5 mg/m2.
Compared with another study of MS-275 in patients with refractory solid tumors and lymphoma (14
), the pharmacokinetic results are consistent, although direct comparison of the PK variables is limited, because MS-275 was administered under nonfasting conditions in that study. In common, however, are the observations of large interpatient variability and dose dependency of systemic exposure as measured by the AUC. This study suggests that the Cmax
increased in a dose-dependent manner. Also similarly, hyperacetylation of histone H3 and H4 was seen in PBMCof patients treated with MS-275, but the amount of hyperacetylation was variable.
Objective responses were only observed in patients receiving the every other week dosing schedule, but the numbers of patients enrolled were too small to determine whether this dosing regimen was truly more efficacious. There did not appear to be any relationship between dose or pharmacokinetic variables and confirmed response or prolonged stabilization of disease. The patient with the metastatic melanoma who experienced a confirmed partial response and who remains on study for >5 years, in fact, was treated at the lowest dose level and required a 25% dose reduction for grade 2 thrombocytopenia after cycle 2. It is indeed intriguing that responses were noted at the lower doses and on the every other week schedule, suggesting that maximal doses and frequent dosing are not required with MS-275, and increasing the attractiveness of such a regimen to patients as well. In other studies of MS-275 in patients with a variety of refractory solid tumors, there appears to be a trend that responding patients were often treated in the lower-dose cohorts, thus suggesting that dose escalation of this agent may not be necessary (NCI Advisory Panel, November 2005).
Overall, MS-275 was well tolerated when administered every other week at doses up to 6 mg/m2 and at doses up to 4 mg/m2 weekly for 3 weeks every 28 days. Either dosing schedule of MS-275 could be considered for further disease-directed studies and should be determined based on individual goals of the studies, biological correlative findings, and the pharmacologic rationale for combinations of agents considered. Based on a single patient in this study who has received MS-275 for 5 years on the every other week schedule, it appears to be well tolerated in chronic, long-term use to date. Further longitudinal observations in additional patients will be critical to follow over time.