Oncolytic measles virus has promising antitumor activity and is being investigated as an experimental cancer therapeutic in various Phase I clinical trials [21
]. In athymic mice bearing intraperitoneal ovarian cancer, we showed that repeat dosing of a recombinant oncolytic measles virus expressing the soluble carcinoembryonic antigen (MV-CEA) virus resulted in significant extension in survival of mice [19
]. A Phase I clinical trial in which MV-CEA was given i.p. to patients with recurrent ovarian cancer was recently completed [23
]. Dose limiting toxicity was not reached in this dose escalation trial in which 6 cycles of 103
of MV-CEA was given [23
]. The best objective response was dose-dependent disease stabilization in 14 of 21 patients with a median duration of 92.5 days (range, 54–277 days). Median survival of patients on study was 12.15 months (range, 1.3-38.4 months).
To improve therapeutic outcome using MV therapy, we focused on maximizing delivery of the virus by using tumor homing cell carriers. We evaluated numerous cell types, including IL-2 expanded T cells [38
], CD14+ derived monocytes [39
], but chose adipose tissue derived MSC due to their ability to be infected by MV and co-localization with ovarian tumors. In contrast to naked virus, these MV infected MSC significantly extended the survival of measles immune tumor bearing mice [33
]. There are also well established standard operating procedures in place for harvest, isolation, culture, expansion and banking of adipose tissue derived MSC in the Human Cell Therapy Laboratory of our institute (A. Dietz). The feasibility and safety of using autologous or allogeneic MSC in humans has been confirmed in a number of phase II and III studies. For example, infusion of autologous MSC in patients with ischemic stroke (2 doses, 5×107
cells/dose, intravenous) or multiple sclerosis and amyotrophic lateral sclerosis (6×107
cells intrathecally or intravenously) or allogeneic MSC to patients with myocardial infarction (0.5 to 5 ×106
/kg intravenous) showed that autologous or allogeneic MSC can be safely administered and tolerated in humans [40
A potential concern with MSC administration is that the cells could be tumorigenic or promote the growth of tumors. Since MSC used in our study will be infected by a replicating virus, the cells typically die within 5 days from virus infection [33
]. In safety studies using athymic mice given a cocktail of ovarian cancer cells and MSC, we showed here that the survival of mice was not significantly different from mice that received tumor cells only. Immunocompromised beige SCID mice were also implanted with large numbers of MSC (4×107
cells/25 gram mouse, 1.6×109
/kg) that had been deliberately passaged in culture for more than 25 passages, representing cell expansion to more than 1012
cells. No tumor growth was seen in mice given MSC subcutaneously or i.p. when necropsied at the end of 90 days, concurring with results from other published reports [43
]. In mice that received large numbers of MSC i.p., a white acellular tissue was seen in the omentum or pancreas, sometimes as an adhesion to the peritoneal wall or inguinal fat pad. Histological analysis using H&E staining and Masson-Trichrome staining revealed that these deposits were acellular connective tissue composed of collagen (positive staining for collagen with Masson-Trichrome stain). The significance of the ‘connective tissue’ is not known, but since MSC home to the omentum post i.p. delivery, we suspect that the deposit represents remnants of the extracellular matrix from the large numbers of cells that have homed to the omentum and died eventually.
Results of this study confirmed that it is highly feasible to isolate and expand MSC from ovarian cancer patients, can be infected by MV, home and deliver the virus to tumor nodules in athymic mice. For our clinical trial, we plan to use autologous patient derived MSC to deliver oncolytic MV in this first-in-human study. The MSCs have a population doubling time that is comparable to MSC from healthy donors and we aim to expand and bank sufficient cells for repeat cycles of treatment (aiming to bank >109
cells per patient). One puzzling finding in our study was the abnormal karyotype of trisomy 20 in the expanded MSC from two of the nine patients. In total, we have tested the karyotype in more than 20 individuals with a variety of diseases including amyotropic lateral sclerosis, Type I diabetes, critical limb ischemia without detection of karyotypic abnormality. Thus, this finding of trisomy 20 in MSC of two ovarian cancer patients is puzzling. Since we did not test the MSC karyotype at harvest, it is not clear if the cells acquired trisomy 20 during expansion in culture. The three instances of viable trisomy are trisomy 21 (Down’s syndrome), trisomy 13 (Patau syndrome) and trisomy 18 (Edward’s syndrome). Complete trisomy 20 is not viable but trisomy 20 mosaicism is estimated to occur in 1 out of 7,000 pregnancies detected prenatally from amniocentesis or chorionic villus sampling [44
]. However, the outcome is normal in 90-93% of prenatally diagnosed cases. Trisomy 20 were noted in non-cancer smokers and mine workers where it was shown they universally harbor increased numbers of abnormal cells within their airway epithelium [45
]. Approximately 50% of ovarian tumors and cell lines have copy number increases in 20q [47
], genes amplified therein, including EEF1A2, may play a central role in the pathogenesis of sporadic and hereditary ovarian carcinoma [46
]. Currently, the significance of trisomy 20 in the expanded MSC of the ovarian cancer patients is not clear but will be the subject of further investigations. In the event that the expanded MSC from the clinical trial fail to meet the release criteria of normal karyotype, the patient will not receive MSC/MV therapy but virus alone.
In this study, we also explored the feasibility of using virus-preinfected cells as the frozen cell bank. In vitro assays indicated that the virus infected cells will proceed to express the viral genes and progeny post thaw from cryopreservation in liquid nitrogen. As shown in this report, actively growing MV-infected MSC and MV infected MSC frozen/thawed cells have comparable antitumor activities. However, since 109
MV was well tolerated in our Phase I trial with preliminary signs of activity, we are proposing to deliver the cocktail of infected cells with the 109
of virus inoculum. Hence, our planned delivery strategy will involve thawing of certified MSC, mixing with our GMP grade virus on the day of treatment, a 5 minute low speed centrifugation, incubation for 2 hours at 37°C, followed by infusion of the cell-virus mixture into the patient by a catheter in the peritoneal cavity. In addition to MSC, other cells types are being developed as carriers of proteins or viruses [9
]. Neural stem cells are being evaluated as carriers of oncolytic adenovirus for glioma therapy [48
] and cytokine induced killer cells for delivery of oncolytic vaccinia virus [50
]. It is hoped that by using cell carriers that home to tumors in conjunction with the replication virotherapy, we could significantly amplify the initial payload of cytotoxic agent to achieve significant improvements in the outcome of cancer therapy.