New Horizons For Brain Tumor Management
January 1, 1998
New Horizons For Brain Tumor Management
ABSTRACTS & COMMENTARY
Synopsis: Brain tumors remain a major challenge because long-term survival is still uncommon. Local recurrence and progression have lethal consequences, and developing novel strategies must focus on local control. Two such novel therapies include gene transfer and intratumor infusion of protein-toxin conjugates. Recent publications have indicated that there is hope on the horizon for both of these approaches, but more work is needed.
Sources: Ram Z, et al. Nature Med 1997;3:1354-1361; Laske DW, et al. Nature Med 1997;3:1362-1368.
The treatment of brain tumors, particularly recurrent glioblastomas, remains a major and, as of yet, insurmountable challenge. In general, disease progression is rapid, and survival is most frequently measured in weeks. Unlike the case for most other solid tumors, it is local progression rather than metastatic disease that proves lethal. Accordingly, it is in this situation that strategies for local control are being developed. Two such approaches involving gene therapy were recently reported in Nature Medicine. The first, by Ram and colleagues, used a technique involving the stereotactic deposition of murine cells (in a solution of up to 10 mL at a concentration of between 108 to 109 cells per mL) along a number of spaced needle tracks within the tumor. The murine cells had been transfected with a retrovirus vector that contained the herpes simplex thymidine kinase gene. The rationale for this approach is based upon an original hypothesis by Mooten and colleagues1 substantiated by a number of animal tumor models (for a review, see Ref. 2). in which the locally expressed thymidine kinase sensitizes tumor cells to systemically administered gancyclovir. One week after deposition, patients were treated with intravenous gancyclovir twice daily for 14 days. Tumor specificity is based upon the concept that integration of the retroviral DNA requires mitotic activity, which would be present in proliferating tumors but not in surrounding normal brain tissue. Nineteen brain lesions from 16 patients were treated. The majority of these were recurrent (after surgery, radiation, and chemotherapy) primary brain tumors, but there were a few patients with metastatic melanoma or breast cancer included. One important criterion for entry into the study was size of the tumor. For a number of theoretical and practical reasons, safety and anti-tumor responses by this approach were considered much more likely if the initial tumor volume were small.
Regression of tumor volume by 50% was seen in five lesions from four patients between 1 and 12 weeks after treatment, but these were transient, with all but one patient progressing 5-11 weeks after peak response. A few of the patients had complications with the needle procedures (either the murine cell depositions or frequent biopsies to evaluate for recurrence) including intratumoral hemorrhage in two cases and increase in seizure activity in three. For the most part, these were well controlled medically, although one of the patients with hemorrhage required surgical resection. Additional important outcomes of this trial were the demonstration of the successful transfection of the retroviral DNA into the tumor cells (albeit only within a few cell layers from the needle tracks) and the demonstration of a lack of retroviral DNA outside of the CNS (by sensitive PCR evaluation of peripheral blood).
Laske and colleagues used a different strategy, one which theoretically offers the advantage of greater distribution of the local anti-tumor effect. They infused human transferrin linked with a mutant diphtheria toxin continuously and directly into the tumors of 18 patients with recurrent malignant brain tumors (16 primary brain tumors and 2 metastatic lung carcinomas). Each of these patients had been treated previously with either radiation, chemotherapy, or surgery, and most had previously received all three modalities.
The rationale for the transferrin conjugate is that rapidly dividing cells, including human glioblastoma, medulloblastoma, and breast cancer cell lines, express increased levels of transferrin receptor. In preclinical studies, the transferrin-diphtheria conjugate was shown to efficiently kill cells expressing transferrin receptor (for a review, see Ref. 3).
One, two, or three catheters were placed into the tumor, and the conjugate was infused continuously for up to 17 days using a high-flow microinfusion technique. The trial was basically a phase I analysis of the approach, and evaluation of toxicity and maximum tolerated dose were the primary objectives. Total volume of infusate and total dose of the conjugate were escalated as the trial progressed until toxicity was observed. However, nine of 15 evaluable patients had more than 50% tumor volume regression including two complete responses. The median survival in the group of nine responders (all of whom had malignant gliomas) was 74 weeks, and three were alive at the time of publication (102-142 weeks after the first treatment).
Systemic toxicity was not observed. However, in all three patients treated at the highest dose (1 mcg/mL), local brain toxicity occurred which resulted in neurological deficits. Seizures occurred in four patients during the infusion, but each of these had experienced seizures prior to therapy and each was controlled medically.
COMMENTARY
With regard to the treatment of primary brain tumors, there is both a profound need for new strategies and a unique opportunity. The profound need relates to the failure of established regimens to provide meaningful responses and the dismal survival rate, especially when considering those with recurrent disease. Brain tumors progress locally; distant metastases do not occur. Thus, the opportunity to develop a local strategy, even for recurrent disease, offers the potential for cure. The use of a retrovirus vector makes sense, and choosing the thymidine kinase/gancyclovir approach reflects a novel application of gene therapy. The paper by Ram and colleagues provides the useful information that the use of both murine cells and a retrovirus vector are safe. Also, the transgene can be successfully transfected to tumor cells in situ, and an anti-tumor response can be expected. However, the loss of replicative capacity of the vector prevents distribution of the gene beyond a few cell layers of the needle track, and the experience indicates that increasing the number of needle passes increases the likelihood of complications. Thus, there’s some encouragement, but a definite need to go back to the drawing board to develop methods to increase the distribution of vector within the tumor.
It certainly appears that the amount of tumor response with the thymidine kinase/gancyclovir approach is greater than the distribution of the retrovirus, implying that the bystander effect observed in the animal models may also occur in people.
Although a less specific technique, the Laske approach seems to overcome the distribution question. Direct and continuous instillation of a large protein conjugated with a cell toxin was proven to be both safe and promisingly effective. What remains is to characterize the dose and approach and gain more experience, perhaps even earlier (e.g., as an adjuvant to surgery, before radiation).
References
1. Mooten FL, et al. Cancer Res 1986;46:5276-5281.
2. Ram Z, et al. Cancer Res 1993;53:83-88.
3. Johnson V, et al. J Neurosurg 1989;70:240-248.
Which of the following statements is correct?
a. Local control of primary brain tumors is achievable in 70% of cases with a combination of surgery and radiation therapy.
b. Gene therapy with intratumoral injection of murine cells programmed to secrete a retrovirus vector containing thymidine kinase DNA to be followed by systemic gancycolvir is effective in the majority of cases, but the toxicity has proven too overwhelming to consider this approach as ever being of clinical value.
c. Transferrin conjugated diphtheria toxin can be safely administered, even to larger primary brain tumors, by microinfusion techniques.
d. Transferrin receptors are not present in brain tumor cells but are present in normal brain cells; therefore, transferrin-conjugated diphtheria toxin is metabolized by the normal cells and inactivated by the normal cells but not the tumor cells.
e. The use of xenogeneic cells, such as murine cells, as vehicles for gene transfer has proven to be ineffective because of immunological forces against the transplanted cells.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content