Gene Therapy for Meningeal Carcinomatosis 
ABSTRACT 
Meningeal carcinomatosis occurs in 5% to 20% of all cancer patients(9, 18, 22). Among 
the adult population, most cases of meningeal carcinomatosis are due to breast or lung carcinomas. 
Over recent years the incidence of meningeal carcinomatosis has been reported to increase, perhaps 
since cancer patients survive longer with improved systemic therapy (1,4, 5, 10,15, 17, 18,22,28). 
Patients with meningeal carcinomatosis have an exceedingly poor prognosis. When maximal 
therapy is tolerated (intrathecal methotrexate and whole-brain irradiation) mean survival is limited 
to 6-7 months(7, 18,30) and fewer than 15% of the patients are alive at one year (26). Meningeal 
involvement is not necessarily a premorbid phenomenon and in two thirds of cases, it occurs when 
systemic disease is stable or in complete remission (7,30). 
In an attempt to improve this grim prognosis of patients with meningeal carcinomatosis, we 
have developed a novel approach for the treatment of meningeal carcinomatosis. This approach 
makes use of recombinant DNA technology to transfer a sensitivity gene into the malignant cells 
seeding the leptomeninges. This is achieved by direct intrathecal injection of cells that actively 
produce a retroviral vector carrying the herpes simplex thymidine kinase gene (HStk), which 
sensitizes the cells to the antiviral drug Ganciclovir (GCV). The intrathecally injected producer 
cells and vector particles can circulate in the CSF and infect cells which are actively synthesizing 
DNA. In the subarachnoid space, such cells are predominantly tumor cells. The HS-tk gene is 
incorporated into the genome of tumor cells and results in expression of the protein encoded by the 
gene. The enzymatic interaction between HS-tk and GCV leads to the production of toxic 
triphospates of GCV that interfere with DNA synthesis and result in the death of the tumor cells. 
Since the thymidine kinase enzyme which is normally present in mammalian cells has very low 
affinity for GCV, systemic toxicity related to this mechanism is not expected. This type of in vivo 
gene transfer has several unique features. First, these retroviral-vectors will only integrate into and 
express their genes in cells which are actively synthesizing DNA. Therefore, non-proliferating 
normal brain and spinal cord tissue should not acquire the HS-tk gene and will remain insensitive 
to GCV. Second, all the transduced tumor cells (and retroviral vector producer cells) will be killed 
by GCV treatment reducing the risk of insertional mutagenesis. Third, in this setting, delivery of 
the producer cells and viral vectors will be facilitated by the normally circulating CSF which bathes 
the whole surface of the CNS. 
The proposed clinical trial will evaluate the dynamics of retroviral vectors in the 
subarachnoid space, assess the safety of this approach and evaluate its potential antitumor efficacy. 
Phase A consists of intraventricular injection of 2 x 10 9 HStk producer cells followed by GCV 
treatment 7 days later for two weeks. In phase B 2 x 10 9 HStk producer cells will be injected 
intraventricularly and intrathecally in the lumbar subarachnoid space (total of 4 x 10 9 HStk cells). 
In phase C, intraventricular and intrathecal injection will be given twice prior to GCV 
administration (8 x 10 9 HStk cells). CSF samples will be analyzed for vector titers at multiple time 
points and, if no toxicity is encountered, 10 patients will be treated in phase D of the study with the 
same parameters as in phase C. Indications of antitumor efficacy will include monitoring of 
clinical symptoms, craniospinal MRI evaluation, and CSF analysis for cytology and tumor 
markers. 
This is the first clinical attempt to treat meningeal carcinomatosis by in vivo genetic 
manipulation of the tumor's genome. 
Recombinant DNA Research, Volume 18 
[597] 
