Clinical Trial of MGI-114 in Prostate Cancer

By Neil N. Senzer, MD

This year, it is estimated that approximately 179,300 men in the United States will develop clinical prostate cancer and that approximately 37,000 deaths will be attributable to this malignancy. The incidence rates for prostate cancer increased 65% between 1980 and 1990, in large part due to the aging of the population, greater awareness of prostate cancer, and prostate specific antigen (PSA) screening. However, since 1996 when an estimated 317,000 men were diagnosed, the trend has reversed. This most likely represents a cull effect as prevalent cases are removed from the population. Yet, while often thought of as a "disease that people die with, not of," prostate cancer remains the second most common organ-specific neoplastic cause of death in men.

Unfortunately, in spite of a high probability of response to initial androgen deprivation, all men with metastatic disease will eventually develop hormone-refractory disease (HRPC). With PSA monitoring, a higher percentage of patients are asymptomatic with recurrent metastatic prostate cancer and, although no more than one-third have measurable disease, almost all have rising PSA values. PSA end points were first reported based on data from a study at the Memorial Sloan-Kettering Institute. That data were confirmed by an NCI study which demonstrated that baseline-to-post-treatment declines of 75% in PSA reliably predicted for survival. Given that single-agent chemotherapy regimens have invariably demonstrated less than 30% response rates in patients with HRPC, when used with reservation,1 PSA analysis in conjunction with measurable disease, bone-only disease, and quality-of-life markers allows for a more rapid and, with hope, accurate assessment of responsiveness to new treatments prior to embarking on longer term and expensive phase III survival end-point studies.

MGI-114 (6-hydroxymethylacylfulvene: HMAF), is a semi-synthetic analog derived from the naturally occurring sesquiterpene, illudin S, isolated from mushrooms of the genus Omphalotus. Accumulation in sensitive tumor cell types is rapid and, following metabolism to a reactive intermediate, it is covalently bound to macromolecules. The degree of DNA binding is closely linked to the cytotoxic potential of MGI-114. DNA synthesis is rapidly and potently inhibited causing cell cycle arrest in S phase. DNA strand breaks and adducts are not readily repaired. Induction and progression of apoptosis is time and concentration-dependent and was demonstrated to decrease with androgen dependency in six human prostate cancer cell lines.2 Of note, MGI-114-induced apoptosis is not affected by caspase-3 inhibitors but is blocked by the broad-spectrum caspase inhibitor Z-VAD-fmk. MGI-114 is effective against both MDR1/gp 170- and MRP/gp 180-positive tumor xenografts3,4 and demonstrates cytotoxicity independent of p53 and p21 status.5 All tested ERCC1-6 (excision repair cross complementing) deficient cell lines retain sensitivity to acylfulvene analogues in contrast to other drugs that covalently bind to DNA which show no more than a two-fold greater sensitivity to ERCC2 or ERCC3 helicase deficient cells.6 Previous studies have demonstrated that ERCC1 upregulation may contribute to the drug resistance observed against standard DNA-binding anticancer drugs.7 Neither ERCC2 nor ERCC3 are known to be upregulated and the apparent requirement for their action in the repair of MGI-114-induced DNA lesions may contribute to the maintenance of drug sensitivity in drug-resistant tumor cell lines.

MGI-114 has demonstrated substantial antitumor activity in two human prostate tumor xenograft models. At the maximum tolerated dose (MTD), 10 of 10 mice showed partial tumor shrinkage (mean shrinkage, 72.5%) in the PC-3 prostate xenograft model. In the DU-145 prostate xenograft model five of eight animals had a complete response at the MTD. Animals that received one-half MTD showed mean rumor shrinkage of 35% with maximum shrinkage (96%) occurring on day 11. At maximally tolerated doses, daily dosing for five days produced significantly better antitumor responses than single doses with responses better than or equal to other multiple dose administration schedules in different xenograft models.

Dosage administrations in phase I studies include five minutes, 30 minutes, and one hour intravenous infusions. All have been daily ´ 5 and repeated every 28 days. The greatest experience has been with the five-minute infusion. The most common drug-related toxicities reported with this schedule are nausea (78%), vomiting (41%), fatigue (50%), facial erythema (41%), and lymphocytopenia (56%). Grade 3,4 nonhematologic toxicities include nausea and vomiting, fatigue, hyperglycemia, and increased alkaline phosphatase. Myelosuppression has been consistently observed at 14.15 mg/m2. Patients with compromised renal function may be at risk for developing renal tubular acidosis. The recommended dose for phase II evaluation was 10.6 mg/m2 over five minutes.

Preliminary results of a Phase II Trial of MGI-114 in Patients with Hormone-Refractory Prostate Cancer were presented at the 35th Annual American Society of Clinical Oncology meeting.8 The study was designed to assess the single-agent antitumor activity of MGI-114 in patients with stage D2 (metastatic) HRPC using a PSA surrogate. Secondary objectives were: 1) to determine the objective response rate in patients with measurable disease (excluding bone); 2) to determine the duration of response and time-to-progression; and 3) to describe and quantify drug related toxicity. MGI-114 at 10.6 mg/m2, daily ´ 5, every 28 days was administered in a five-minute infusion. A PICC line and prophylactic 5-HT3/steroid antiemetics were required. At the time of the report, 31 patients were registered with documented disease progression having received no more than two hormone-suppression regimens. Twenty-one of the patients (mean age, 68; range, 52-84) are evaluable. The mean pretreatment PSA is 306 ng/mL (range, 11.6-2294 ng/mL). At study entry, patients had ECOG scores of 0 (6 patients), 1 (14 patients), and 2 (1 patient). Prior endocrine treatments include orchiectomy in eight, LH/RH agonist in 14, and antiandrogen (discontinued 4-6 weeks prior to initiation of therapy) in 18 patients. Six patients received primary radiation therapy and nine, palliative. Sixteen patients exhibited metastatic disease to bone, six to lymph node, two to liver, and three in multiple sites.

Grade 3 asthenia and nausea occurred in three patients each. Grade 3 thrombocytopenia and neutropenia each occurred in six patients, with prolonged thrombocytopenia emerging as the primary cause of treatment delay. PSA responses included three patients with greater than a 50% reduction for a median 23 weeks duration (range, 2-26+) and 13 patients with stable PSA for a median 13 weeks duration (range, 9-33+) for an overall response rate of 76%. To date, one patient has experienced a complete soft-tissue response (retroperitoneal lymph nodes) and, a second patient, a partial soft-tissue response (iliac nodes). The study is continuing with a targeted accrual of 30 evaluable patients.

MGI-114 is reasonably well tolerated in this "chemo-naïve" group of patients. Nausea and vomiting, usually emerging on the fourth to fifth day of infusion, is controlled with a prophylactic HT3/steroid combination, which should be continued orally. MGI-114 appears to have sufficient clinical activity in patients with hormone-refractory prostate cancer to warrant further assessment, especially in combination with other compounds and/or radiation. Britten and colleagues9 have recently described a supra-additive interaction between MGI-114 and CPT-11 in the HT29 human colon tumor xenograft. Although there is no in vitro evidence that a topoisomerase is required for nucleotide excision repair, mismatch repair, or repair by DNA synthesis, inhibiting topoisomerase could alter the level of supercoiling in a chromosomal domain, which could lead to the recruitment of other enzymes acting on the damaged site. In the MVI-522 human lung xenograft model, Mangold10 was able to show significant enhancement of response to the combination of MGI-114 with the topoisomerase inhibitor topotecan and paclitaxel. DNA helicase inhibitors (e.g., CI-958, distamycin, and mithramycin) have also been shown to act synergistically with MGI-114 in a series of pediatric tumor cell lines.11 Finally, given the recognition of the role of transcription coupled nucleotide excision repair in radiation ROI mediated damage and the role of the Ku80 helicase in homology dependent DNA double-strand break repair, the potential for coordinated delivery of MGI-114 and radiation is a reasonable combination to investigate. (Dr. Senzer is Director of Radiation Oncology Research, Co-Director Urologic Oncology, Sammons Cancer Center, Dallas, TX.)


1. Dawson NA. Response criteria in prostatic carcinoma. Sem Oncology 1999;26:174-184.

2. Woynarowska BA, et al. Proc Am Assoc Cancer Res 1999;40:488.

3. Kelner MJ,et al. Proc Am Assoc Cancer Res 1999;40:300.

4. Kelner MJ, et al. Eur J Cancer 1998;34:908.

5. Izbicka E, et al. Proc Am Assoc Cancer Res 1999;40:300.

6. Kelner MJ, et al. Cancer Res 1995;55:4936-4940.

7. Dabholker M, et al. J Natl Cancer Inst 1992;84: 1512-1517.

8. Senzer NN, et al. ASCO Program/Proceedings 1999;18:321a.

9. Britten, et al. Cancer Res 1999;59:1049-1053.

10. 10th NCI-EORTC Symposium on New Drugs in Cancer Therapy, 1998.

11. Moore R, et al. Proc Am Assoc Cancer Res 1999;40: 590-591.