Intrathecal gene therapy for treatment of leptomeningeal carcinomatosis

Abstract

Leptomeningeal carcinomatosis occurs occasionally in patients with solid malignancies and carries a poor prognosis despite treatment with systemic chemotherapy and/or radiotherapy. We describe the case of a 43 year old man who presented with leptomeningeal carcinomatosis secondary to malignant melanoma. The patient received intraventricular delivery of NIH3T3 producer cells expressing the thymidine kinase (HSV-Tk1) gene via a retroviral vector followed by intravenous ganciclovir. He experienced abrupt and severe meningeal irritation and hyperpyrexia immediately after injection of the producer cells into the ventricular CSF. Vector producer cells (VPC) survived and were detected by NeoR marker gene expression in the CSF for a week, until a single dose of ganciclovir (GCV) was followed by a decline in the copy number of the NeoR marker gene to undetectable levels over 24 h. This decline upon introduction of ganciclovir suggests effective distribution of ganciclovir to producer cells bearing the HSV-Tk gene. The patient survived 9 months after treatment. Side-effects from the treatment included acute hyperpyrexia which was short-lived and medically manageable.

Introduction

Leptomeningeal carcinomatosis (LM) is a debilitating and rapidly progressing manifestation of solid tumors whose incidence is increasing [1–5]. Although LM may develop in any cancer, breast cancer (half of cases), lung cancer, and melanoma are the most common primaries [6–10]. Even when maximal standard therapy is tolerated (intrathecal methotrexate and involved field irradiation), mean survival is only 5–6 months and 1-year survival is less than 15% [3, 6]. Here we report the case of a man who presented with LM secondary to malignant melanoma and was treated with intrathecal gene therapy with retroviral delivery of the thymidine kinase (HSV-Tk1) gene followed by intravenous ganciclovir.

Case report

History and examination

This 43 year old man was diagnosed with melanoma in 1989. The same year he underwent surgery and irradiation to treat an extradural and intradural metastasis that extended from C5 to T2 spinal levels. 5 years later he underwent follow-up evaluation with lumbar puncture and spinal MRI. Melanomatous cells were found in the CSF at that time. Spinal MRI demonstrated contrast enhancement within the thoracic and lumbar intrathecal space (Fig. 1). The patient was informed of treatment options. In August 1994, he enrolled in a clinical trial: Intrathecal Gene Therapy for the Treatment of Leptomeningeal Carcinomatosis (GTI0108), a Phase I/II Study. The study was approved by the National Institutes of Health Recombinant DNA Advisory Committee and the Institutional Review Board of the NINDS. Informed consent was obtained from the patient, who was evaluated and treated at the Clinical Research Center of the National Institutes of Health, Bethesda, Maryland.

Fig. 1.

Before treatment began, T1-weighted, 0.5 T, MRI of the lumbar spine without (a) and with (b) gadolinium contrast demonstrated enhancing leptomeningeal tumor (arrowheads)

Treatment phase

The patient underwent three procedures (1) surgery for placement of a catheter in the right lateral ventricle and attachment to an Ommaya reservoir; (2) insertion of a lumbar catheter for sampling of cerebrospinal fluid for 7 days after cell injection; and on the second post-operative day; (3) injection of 10 ml of 1 × 10⁹ GlTklSvNa. 7 vector producer cells over 10 min into the ventricular CSF via the Ommaya reservoir. “GlTklSvNa. 7 is a retroviral vector derived from the Moloney murine leukemia virus (MoMLV)” [11]. This vector contains herpes simplex virus thymidine kinase (HSV-Tk1) gene cDNA that is transcribed downstream from the viral LTR (long terminal repeat) and contains the bacterial neomycin resistance (NeoR) marker gene transcribed from an internal SV40 (simian virus 40) early promoter (LTR-HSV-Tk1-SV-NeoR-LTR) in the G l vector backbone (Genetic Therapy Inc., Gaithersburg, MD). This G1-based vector has been modified for increased safety by alteration of the retroviral GAG gene which encodes the viral matrix, capsid and nucleoproteins resulting in the elimination of viral sequences needed for formation of a replication-competent virus [11]. “This has been shown to decrease the potential for helper virus production from the producer cells which contain the vector. No replication-competent virus has been detected in vectors administered to patients or following administration of the vector to animals or humans [11]. The vector is packaged by the amphotropic retroviral-vector producer cell line PA317, which is derived from NIH3T3 cells and has a titer of 1×10⁴ to 1×10⁵ colony-forming units/ml of NIH 3T3 cells” [11].

Intravenous administration of ganciclovir (Cytovene, Syntex Corporation, Palo Alto, CA) 5 mg/kg, twice daily [11], began 7 days after injection of vector producer cells and continued for 14 days. On the day of ganciclovir infusion and for the next 7 days thereafter the patient underwent twice-daily sampling of CSF from the lumbar drain and from the Ommaya reservoir. DNA was isolated from CSF specimens collected during the study and was evaluated by PCR for the vector NeoR marker gene in the producer cell, the genome copy number reflecting producer cell number in the subarachnoid space (Fig. 2). Serial MR scans were taken before ganciclovir administration and on days 7 and 14 of ganciclovir treatment.

Fig. 2.

NeoR vector expression by producer cells persisted for 7 days in the CSF. Titers, expressed as vector genome copy number per mL of CSF, fell to zero by 24 h after administration of IV ganciclovir

Post treatment course

Within 5 min of receiving the vector producer cells the patient developed acute meningeal irritation and pyrexia (Tmax 40.6°C), which was managed successfully with Dexamethasone and Tylenol. On day 12 of ganciclovir treatment, the patient developed mild weakness of his lower extremities of insidious onset that appeared to be disease-related. Weakness did not prevent ambulation and he was discharged from the hospital 10 days later. One month after discharge he developed progressive lower extremity weakness and became non-ambulatory. He did not return for follow up evaluation. His family reported that he survived for 9 months after treatment. A postmortem examination was not performed.

Discussion

It is noteworthy in this study that vector producer cells (VPC) survived and constantly expressed retroviral gene product in the CSF for a week, until a single dose of GCV resulted in a reduction in vector DNA over the next 24 h. The meningeal irritation observed in this patient 5 min after infusion of vector producer cells apparently resulted from an acute inflammatory response to the vector producer cells by the meninges [12]. Acute hyperpyrexia likely resulted from the large protein load that was delivered into the CSF by intraventricular injection of 1 billion murine cells. The production of retroviral vectors was followed by using a semi-quantitative PCR method to measure the NeoR vector genome copy number per mL in the lumbar CSF of the patient. Retroviral vector DNA persisted for 1 week after producer cell injection (Fig. 2). Following ganciclovir administration, the NeoR marker copy number in lumbar CSF declined to zero over 24 h. This rapid reduction in retroviral marker upon introduction of ganciclovir suggests effective distribution of ganciclovir to vector producer cells, immediate cessation of vector production, and fairly rapid clearance of viral DNA from the CSF.

Despite an apparent cytotoxic effect of ganciclovir on the vector producer cells, no significant changes were reported in MRI scans or CSF cytology following treatment. Treatment was therefore insufficient to significantly reduce the extensive leptomeningeal tumor load or number of tumor cells in the CSF.

Leptomeningeal carcinomatosis is reported in 5% of patients with solid tumours [13]. The clinical course after diagnosis is typically one of progressive neurological decline followed by fatal outcome within 6–12 months [11]. Management of this condition is essentially palliative. The conventional treatment regimen of chemotherapy with or without craniospinal irradiation has significant toxicity and limited efficacy [14]. In light of the shortcomings of conventional treatment, gene therapy was considered to be a method that could potentially improve on standard care.

The possible role of gene therapy in the treatment of diseases of the CNS is an area of intense scientific and medical interest. The central nervous system may be more amenable to gene therapy than other organs because it has reduced immune response to viral vectors compared to other organs [2, 15, 16]. The effectiveness of gene therapy may be augmented by combining it with other therapies; head and neck cancer and other tumors may become more susceptible to the effects of irradiation after exposure to gene therapy [8, 17, 18]. Various clinical trials have shown that transfer of the HSV-Tk gene followed by ganciclovir is well tolerated in brain tumor patients [16, 18, 19]. The treatment of this patient provided unique observations about intrathecal gene therapy that deserve reporting, even though the use of these producer cells as an in situ source of retroviral vectors to transduce tumor cells with the HSV-Tk suicide gene is no longer approved by the FDA for investigational gene therapy trials. The patient received a novel treatment intervention using cells that actively produced a retroviral vector carrying the herpes simplex thymidine kinase gene (HSV-Tk1), which sensitizes transfected tumor cells to the antiviral drug ganciclovir. The retroviral vector affects only actively dividing cells because it does not penetrate the nuclear membrane and can only enter the nucleus during mitosis when the nuclear membrane breaks down. The producer cells were intended to circulate in the CSF and release the retroviral vector to dividing tumor cells. This method would target tumor cells because spinal cord neurons do not divide in adults and are protected from transfection by this vector. In the subarachnoid space and on the leptomeninges of this patient, actively dividing cells were predominantly tumor cells. In tumor cells that become transfected, the enzymatic interaction between HSV-Tk1 and ganciclovir would lead to the production of toxic triphosphate of ganciclovir that would interrupt DNA synthesis and kill tumor cells. The toxic triphosphate of ganciclovir could also reach adjacent tumor cells through cellular contacts and kill these cells, producing the so-called bystander effect. Ganciclovir would kill transduced tumor cells and retroviral vector producer cells, reducing the risk of insertional mutagenesis [11]. This approach would have minimal systemic toxicity because mammalian thymidine kinase has very low affinity for ganciclovir.

The treatment employed in this protocol is believed to be the only case of intrathecal gene therapy for LM patients. The rationale of this treatment was based on promising results, in terms of prolongation of survival and elimination of infiltrative tumor, which had emerged from studies on animal models of LM treated with the HSV-Tk/ganciclovir combination [5, 8]. In pre-clinical experiments the retroviral vector that was used in this patient had been able to selectively transduce mitotically active cells while sparing adjacent non-dividing cells. Transfection of every tumor cell was not essential for tumor elimination in animal models because regression of tumors could occur with this treatment regimen despite limited transduction, a phenomenon that was attributed to a “bystander” effect [18]. Despite showing anti-tumor effect in preclinical studies, the treatment did not significantly reduce tumor load or affect the natural history of leptomeningeal carcinomatosis in this patient. This clinical trial in one patient adds to the public record information about human intrathecal gene therapy.

This experimental paradigm was previously evaluated with intraparenchymal delivery in glioma patients in a Phase III trial, and showed no benefit [19, 20]. Histological analysis in another study of glioma patients suggested that this therapy was ineffective because producer cells and vector did not disseminate from the infusion site and failed to transduce tumor cells with the Tk gene [21].

The case report describes an unsuccessful attempt to treat leptomeningeal carcinomatosis with gene therapy. Examination of this treatment failure can inform future trials of gene therapy using other vectors. One explanation for lack of efficacy was that HSV-Tk gene delivery to the tumor was inhomogeneous and insufficient to catalyze conversion of enough ganciclovir to ganciclovir triphosphate throughout the tumor and eradicate the tumor mass. This theory is consistent with findings in a recent report of herpes simplex virus-thymidine kinase treatment of leptomeningeal glioma in rats. Mesenchymal stem cells transduced with the herpes simplex virus-thymidine kinase gene were injected into the cisterna magna 24 h after tumor cells had been injected there, followed by systemic ganciclovir for 10 days [22]. The stem cells were designed to synthesize ganciclovir to ganciclovir triphosphate which would act on adjacent tumor cells through the bystander effect. This treatment method prolonged survival in the treated group only 25% longer than in the control group. Although transduced mesenchymal stem cells migrated into the leptomeningeal tumor and subsequent treatment with ganciclovir reduced tumor size, subarachnoid tumor did not resolve completely in any animal. The same group in an earlier study reported that the ratio of stem cells to tumor cells must exceed 1:32 to successfully eradicate brain tumors that originated from implanted tumor cells that were co-injected with mesenchymal stem cells transduced with the herpes simplex virus-thymidine kinase gene [23]. The lack of efficacy of the HSV-tk/ganciclovir system against human leptomeningeal carcinomatosis in our case report is likely related to the inability of viral particles to penetrate deeply enough from the CSF to distribute throughout the tumor, which would limit its cytotoxic effects on dividing cells. Lentivirus might be an alternative approach for HSV-Tk/ganciclovir gene therapy of leptomeningeal carcinomatosis because it can infect dividing and non-dividing cells, spread from a host cell to infect neighboring cells, and penetrate deeper into the tumor than was possible with the vector used in this study [24].