Moore, D., Donnelly, J., McGuire, W.P., Almadrones, L., Cella, D.F., Herzog, T.J., & Waggoner, S.E. (2003). Limited access trial using amifostine for protection against cisplatin and three hour paclitaxel-induced neurotoxicity: A phase II study of the Gynecologic Oncology Group. Journal of Clinical Oncology, 21, 4207–4213.

The researchers' aim was to determine the proportion of women who experience significant treatment-induced peripheral neuropathy.

Women with gynecologic cancer received combination chemotherapy consisting of cisplatin and paclitaxel via IV 175 mg/m² over three hours followed by amifostine 740 mg/m² and cisplatin 75 mg/m² administered over 90 minutes beginning 15 minutes after amifostine administration.

Twenty-nine women enrolled in the limited-access study and 21 completed all six cycles of therapy plus the three-month evaluations. The eligibility criteria for this phase II study included having ovarian cancer and primary peritoneal, fallopian tube, endometrial, cervical, or uterine sarcoma. Exclusion criteria include prior chemotherapy or radiation therapy and having a history of neuropathy.

Measures were conducted at baseline, before each chemotherapy cycle, and three months after completion of treatment. Chemotherapy-induced peripheral neuropathy (CIPN) was measured using the National Cancer Institute's Common Terminology Criteria for Adverse Events (NCI-CTCAE) toxicity scale, the neurotoxicity subscale of the Functional Assessment of Cancer Therapy–Gynecologic Oncology Group (FACT-GOG), and vibration perception threshold (VPT) was tested using the Vibratron II device.

Four of the 27 assessable patients experienced dose-limiting toxicity (greater than grade 2 CIPN) as measured by clinical assessment and the NCI-CTCAE. The number of neuropathic events exceeded the predetermined threshold level for a second stage of accrual and the study was closed. The level of activity for amifostine was insufficient to warrant additional study in a phase III trial.

The study contained a relatively small sample size (29 participants) and used a one-group design. Frequent evaluations of CIPN were performed by different professionals, but a detection bias may have been present. Although VPT testing is considered to be the gold standard, it can be inaccurate if the same digits are not used at each assessment. Low interobserver agreement (46%) was found using the NCI-CTCAE.