The mda-7 gene (approved gene symbol IL24) is a novel tumor suppressor gene with tumor-apoptotic and immune-activating properties. We completed a Phase I dose-escalation clinical trial, in which a nonreplicating adenoviral construct expressing the mda-7 transgene (INGN 241; Ad-mda7) was administered intratumorally to 22 patients with advanced cancer. Excised tumors were evaluated for vector-specific DNA and RNA, transgenic MDA-7 expression, and biological effects. Successful gene transfer as assessed by DNA- and RT-PCR was demonstrated in 100% of patients evaluated. DNA analyses demonstrated a dose-dependent penetration of INGN 241 (up to 4 x 10(8) copies/mug DNA at the 2 x 10(12) vp dose). A parallel distribution of vector DNA, vector RNA, MDA-7 protein expression, and apoptosis induction was observed in all tumors, with signals decreasing with distance away from the injection site. Additional evidence for bioactivity of INGN 241 was illustrated via regulation of the MDA-7 target genes beta-catenin, iNOS, and CD31. Transient increases (up to 20-fold) of serum IL-6, IL-10, and TNF-alpha were observed. Significantly higher elevations of IL-6 and TNF-alpha were observed in patients who responded clinically to INGN 241. Patients also showed marked increases of CD3+CD8+ T cells posttreatment, suggesting that INGN 241 increased systemic TH1 cytokine production and mobilized CD8+ T cells. Intratumoral delivery of INGN 241 induced apoptosis in a large volume of tumor and elicited tumor-regulatory and immune-activating events that are consistent with the preclinical features of MDA-7/IL-24.

Purpose: To determine maximum tolerated dose of CI-994, a novel oral histone deacetylase inhibitor, in combination with carboplatin and paclitaxel in patients with advanced solid tumors.Patients and Methods: Patients with advanced solid tumors who had received two or fewer prior chemotherapy regimens were eligible for trial. Five cohorts of patients were treated with escalating doses (4-6 mg/m(2)) and alternative schedules (7 days or 14 days) of CI-994. Dose escalation of paclitaxel was performed to achieve tolerability of CI-994 with a paclitaxel dose of 225 mg/m(2) when administered in combination with carboplatin. Pharmacokinetic assessment of CI-994 was performed by using liquid chromatography/mass spectrometry. Histone deacetylation inhibition was determined by Western blot analysis.Results: A total of 30 patients (median age 58 years) were entered into five treatment cohorts. Maximum tolerated dose of CI-994 was determined to be 4 mg/m(2) administered for 7 consecutive days following paclitaxel at a dose of 225 mg/m(2) and carboplatin at an area under the curve (AUC) of 6 every 21 days. Neutropenia, thrombocytopenia, and grade 3 respiratory insufficiency limited further dose escalation of CI-994. Pharmacokinetics showed that CI-994 absorption and disposition were unaffected by carboplatin and paclitaxel coadministration. Association between histone H3 acetylation levels and disease response was suggested. A subset of patients with lymphocyte H3 acetylation levels at least 1.5-fold times baseline all achieved either a clinical response or stable disease. All evaluable patients with progressive disease (PD) had H3 acetylation levels < 1.5-fold times baseline. Twenty-four of the 30 patients received greater than one cycle of treatment. Five of these patients achieved a partial response (3 nonsmall cell lung cancer, 1 colorectal cancer, and 1 unknown primary) and 2 patients achieved a complete response (esophageal and bladder cancer).Conclusion: The combination of CI-994 at a dose of 4 mg/m(2) administered orally for 7 consecutive days can be safely coadministered with paclitaxel at a dose of 225 mg/m(2) and carboplatin at an AUC of 6 on day 1 of a 21-day cycle. Evidence of antitumor activity is suggested and may correlate with histone modulation.

One emerging approach to improve the treatment of drug-resistant cancers involves targeting of therapeutic genes by adenoviral delivery to tumor cells. Resistance to cytotoxic agents commonly occurs because the cytotoxic is an external agent that depends upon cellular exposure to achieve its affects. Decreases in uptake or retention by the cancer cell result in suboptimal exposure and reduced effect. Recent understanding of basic cellular processes in normal cells and how they go awry in cancer has yielded new therapeutic opportunities using viral delivery vehicles to target the cancer specific abnormalities (Borst et al., 2001; Nemunaitis and Cunningham, 2002). Persistent investigation into the mechanism(s) of selective replication and associated oncolytic effect of DNA viruses in malignant tissue continues to yield fruitful results. Three replicative DNA viral constructs: adenovirus (Khuri et al., 2000; Nemunaitis et al., 2001b), herpes simples virus (Kucharzuk et al., 1997; Toyozuimi et al., 1999) and vaccinia virus (Gomella et al., 2001; Mastrangelo et al., 2000) are currently being explored clinically. Each of these viruses has demonstrated reasonable tolerability and evidence of activity when administered via local regional approaches, however, none have shown systemic activity. Our own experience with ONYX-015 (an E1b deleted selective replicating adenovirus) illustrates this point. Despite significant clinical activity with intratumoral injection of ONYX 015 (Khuri et al., 2000; Nemunaitis et al., 2001b), intravenous infusion produced no clinical responses despite clear evidence of viral access to malignant tissue. Electron micrographs of tumor obtained from treated patients showed clumps of intranuclear viral particles within malignant tissue, but there was little or no evidence of oncolysis (Nemunaitis et al., 2001a, in press). We theorized that this lack of activity might be due to low concentrations of viral particles actually presenting to malignant tissue because of hepatic

ONYX-015 is an adenovirus that selectively replicates in p53 dysfunctional or mutated malignant cells. We performed a pilot trial to determine the safety and feasibility of treatment with ONYX-015 delivered intravenously in patients with advanced malignancy. One cohort of five patients received ONYX-015 once a week for 6 weeks at a dose of 2 × 10 12 particles per infusion in combination with weekly infusions of irinotecan (CPT11, 125 mg per week) and 5-fluorouracil (5FU, 500 mg per week). A second cohort of five patients received the combination of ONYX-015 at a dose of 2 × 10 11 particles per week for 6 weeks in combination with interleukin 2 (IL 2, 1.1 × 10 6 units daily via subcutaneous injection for 5 days each week for 4 weeks). Toxicity attributable to ONYX-015 was limited to transient fever. All patients demonstrated elevations in neutralizing antibody titers within 4 weeks of the infusion of ONYX-015. Serum levels of IL-6, IL-10, tumor necrosis factor- α , and interferon- γ increased within 6 hours of viral infusion, suggesting immune activation. This response was more pronounced in the cohort of patients who received 2 × 10 12 particles per infusion. Two patients demonstrated uptake of viral particles in malignant tissue by quantitative PCR. Electron microscopy confirmed selective cytoplasmic viral particles within malignant cells but not within adjacent normal tissue in a third patient. In conclusion ONYX-015 can be administered safely in combination with CPT11, 5FU or low-dose IL 2 and is able to access malignant tissue following intravenous infusion. Further investigation of ONYX-015, possibly with agents that may modulate replication activity, or duration of virus survival, is indicated.

Introduction of the herpes simplex type I thymidine kinase (HSV-TK) gene into tumor tissue, followed by ganciclovir, initiates a phosphorylation cascade that induces formation of a toxic ganciclovir triphosphate. Animal trials suggest that this ganciclovir triphosphate has antitumor activity. Here we report application of the HSV-TK transfection approach using a retroviral construct. Sixteen patients (median age 61.5 years) with refractory carcinoma (13 melanoma, 1 breast cancer, 1 nonsmall-cell lung cancer, and 1 osteogenic sarcoma) received intratumoral injection of HSV-TK retroviral vector at escalating doses (0.2 x 10(7) cfu per injection x 5 daily doses) and we evaluated them for toxicity and activity. We observed grade III pain associated with cellulitis in one patient following injection. Analysis of blood samples drawn between 3 and 28 weeks from 14 patients for replication-competent retrovirus by PCR analysis of the amphotrophic envelope revealed no replication-competent retrovirus. We injected 21 lesions. We identified no tumor responses of the injected lesions. Of 13 patients with advanced melanoma, 6 survived over one year. Thus, injection of retroviral delivered HSV-TK in patients with refractory cancer was well-tolerated.

Although genetically engineered adenoviruses hold promise for the treatment of cancer, clinical trial reports have utilized intratumoral injection to date. To determine the feasibility of intravenous delivery of ONYX-015, an E1B-55kD gene-deleted replication selective adenovirus with demonstrated clinical safety and antitumoral activity following intratumoral injection, we performed a clinical trial in patients with metastatic solid tumors. ONYX-015 was infused intravenously at escalating doses of 2 × 1010 to 2 × 1013 particles via weekly infusion within 21-day cycles in 10 patients with advanced carcinoma metastatic to the lung. No dose-limiting toxicity was identified. Mild to moderate fever, rigors and a dose-dependent transient transaminitis were the most common adverse events. Neutralizing antibody titers significantly increased within 3 weeks in all patients. IL-6, γ-IFN, TNF-α and IL-10 increased within 24 h following treatment. Evidence of viral replication was detectable in three of four patients receiving ONYX-015 at doses ⩾2 × 1012 particles and intratumoral replication was confirmed in one patient. In conclusion, intravenous infusion of ONYX-015 was well tolerated at doses up to 2 × 1013 particles and infection of metastatic pulmonary sites with subsequent intratumoral viral replication was seen. The intravenous administration of genetically altered adenovirus is a feasible approach.

Collins Building, 5th Floor, Dallas, Texas, USA T he p53 gene product is responsible for several growth regulatory functions, including the cell’s initial response to a DNA virus infection. Although most investigations have looked at the role of p53 in inducing a G1 growth arrest via the cyclin-dependent-kinase inhibitor p21/WAF1/Cip1, recent evidence suggests that G2 checkpoint mechanics may also be dependent on the p53pathway response. Raj, et al. exploit this finding in their paper detailing the selective viral killing of cells that lack p53 activity with a type 2 non-replicating adeno-associated virus (AAV2). In doing so, the authors may have also cast new light on another current controversy, i.e. the selectivity of E1B-deleted adenoviruses. In the current paper, destruction of the p53-dysfunctional cancer cells was via an apoptotic pathway that did not depend on viral DNA replication or gene expression, but presumably on the structure of the AAV2 genome itself. The AAV2 genome is comprised of single-stranded DNA with inverted terminal repeats (ITRs) forming hairpin structures at each end. The authors postulate that these structures elicit a DNA damage response that induces a G2 phase arrest in all cells. Normal cells (with an intact p53 system) can remain at this point but p53-deficient tumor cells cannot. Failure to sustain G2 arrest triggers an apoptotic pathway so that the end result is selective killing of cells lacking p53 function. This is not the first instance of p53-selective viral cytolysis. Onyx-015 is an E1B 55kD-deleted mutant chimeric human group C adenovirus being developed as a potential cancer therapeutic that targets p53 mutant cancer cells. The E1B-55kD gene product binds and inactivates p53 so that its deletion should confer selective adenoviral replication in p53 dysfunctional cells. Initial studies confirmed this hypothesis; normal cells with intact p53 activity were non-permissive for productive Onyx-015 infection, whereas cancer cells with an impaired p53 pathway were more permissive, resulting in a selective oncolytic effect in vitro. The key point here is that this selective oncolytic effect has been explained on the basis of large differences in viral replication. Using isogenic tumor cell lines that express either wild-type or a dominant negative mutant of p53, Rogulski, et al. observed a three times faster replication rate with Onyx 015 in p53 mutant cells compared to p53 normal cells. This correlated with a proportionately increased cytolysis rate in vitro. Other work from Harada found that temperature-sensitive mutants of p53 had an approximately three-fold slower viral replication rate and a lower viral burst rate when p53 was active. However, later investigations gave more mixed results in that p53 status did not always correlate with viral cytotoxicity.11–13 These divergent results are clouded by differences in multiplicity of infection