1. ¹Cancer Gene Therapy Group, Rational Drug Design Program and Haartman Institute, University of Helsinki, Helsinki, Finland
2. ²Department of Oncology, Helsinki University Central Hospital, Helsinki, Finland
3. ³Department of Obstetrics and Gynecology, Helsinki University Central Hospital, Helsinki, Finland
4. ⁴Pathology/HUSLAB, Helsinki University Central Hospital, Helsinki, Finland
5. ⁵Molecular and Cancer Research Biology Program, University of Helsinki, Helsinki, Finland
6. ⁶Department of Pathology, University of Florida, Jacksonville, FL, USA

Correspondence: Dr A Hemminki, Cancer Gene Therapy Group, Rational Drug Design Program, University of Helsinki, PO Box 63 (Haartmaninkatu 8),Biomedicum Helsinki; 00014 Helsinki, Finland. E-mail: akseli.hemminki@helsinki.fi

Received 30 August 2005; Revised 26 April 2006; Accepted 8 June 2006; Published online 10 August 2006.

Abstract

Conditionally replicating adenoviruses (CRAds) that replicate in tumor but less in normal cells are promising anticancer agents. A major determinant of their potency is their capacity for infecting target cells. The primary receptor for serotype 5 adenovirus (Ad5), the most widely used serotype in gene therapy, is the coxsackie-adenovirus receptor (CAR). CAR is expressed variably and often at low levels in various tumor types including advanced breast cancer. We generated a novel p16/retinoblastoma pathway-dependent CRAd, Ad5.pK7-24, with a polylysine motif in the fiber C-terminus, enabling CAR-independent binding to heparan sulfate proteoglycans (HSPG). Ad5.pK7-24 mediated effective oncolysis of all breast cancer cell lines tested. Further, we utilized noninvasive, fluorescent imaging for analysis of antitumor efficacy in an orthotopic model of advanced hormone refractory breast cancer. A therapeutic benefit was seen following both intratumoral and intravenous delivery. Murine biodistribution similar to Ad5, proven safe in trials, suggests feasibility of clinical safety testing. Interestingly, upregulation of CAR was seen in low-CAR M4A4-LM3 breast cancer cells in vivo, which resulted in better than expected efficacy also with an isogenic CRAd with an unmodified capsid. These results suggest utility of Ad5.pK7-24 and the orthotopic model for further translational studies.