Chemoablation of melanoma with intralesional rose bengal (PV-10)

Abstract

9060 Background: Intralesional rose bengal (PV-10, a sterile 10% solution in saline) can elicit selective ablation of solid tumors and an apparent bystander response in untreated lesions. In phase 1 testing in 20 subjects with AJCC Stage III (19 subjects) or IV (1 subject) melanoma, a single injection of PV-10 into 1–20 lesions led to durable objective response (OR) at 12–24 weeks in 40% of subjects (20% CR + 20% PR by modified RECIST) and locoregional disease control (CR + PR + SD) in 75% of subjects. Untreated bystander lesions achieved an OR in 15% of subjects, and all subjects with an OR of their injected lesions achieved disease control of their bystander lesions. PV-10 was well tolerated, with only one Grade 3 adverse event (photosensitivity reaction). The most common AEs were transient pain at the treatment site (75% of subjects), followed by local inflammation or infection (25%). Methods: Expanded phase 2 testing commenced in late 2007 in up to 80 subjects with measurable Stage III or IV melanoma. After an initial treatment of 1–20 cutaneous, subcutaneous or nodal lesions, new or incompletely responsive lesions can be retreated at weeks 8, 12 or 16, with follow-up to 52 weeks. An additional 1–2 lesions, including visceral lesions, remain untreated for assessment of bystander response. Seven centers in Australia and the USA are enrolling subjects, with completion of enrolment expected by mid-2009. A modified Fleming design allows interim assessment of safety and OR at weeks 4 and 24, respectively, after treatment of the 20th and 40th subjects. The primary end-point is OR of injected lesions in the intent-to-treat population; secondary endpoints include OR of bystander lesions and PFS. Results: Interim safety data for the first 40 subjects treated is comparable to phase 1, with transient mild to moderate locoregional pain, vesicles or edema most common; Grade 3 AEs have been rare (1 case each of vesicles and skin flap necrosis), with no Grade 4 or 5 AEs attributed to PV-10. Interim efficacy for the first 20 subjects is also comparable to that of phase 1 (15% CR + 15% PR); efficacy data for the first 40 subjects will be presented at the meeting. Conclusions: The safety and efficacy profile of intralesional therapy with PV-10 compares favorably with available therapeutic options for this patient population. No significant financial relationships to disclose.