be summarized as follows:
Tumors are metastatic at the time of diagnosis. By Gene therapy has captured the imaginations of bio-the time signs and symptoms appear, most tumors are advanced. In the case of solid tumors, advanced disease medical scientists because of its perceived potential to dramatically improve therapies for the most challeng-is often characterized by the presence of multiple tumor masses, some of which lie quite distant from the pri-ing medical problems. What could be more ''natural,'' after all, than modifying DNA profiles of individuals mary lesion. Thus, while stereoscopic insertion of retroviral vectors that produce tk is quite efficient at de-or their diseased cells so as to render them genetically resistant to pathogenic anomalies? Given this prospect, stroying gliomas in the brains of rats, 1 such procedures can only be expected to destroy the mass that is treated it is hardly surprising that neoplastic diseases have drawn the attention of gene therapists. These mala-directly, and not to impair growth of geographically distant metastatic foci. Thus, such treatments for wide-dies, often inexorably progressive, psychologically and physically debilitating, and fatal, are matched in their spread solid tumors would function as a form of ''genetic surgery,'' reducing tumor mass but not effecting perniciousness only by their prevalence; indeed, they constitute a major cause of death in all developed coun-a cure. Where disseminated hematologic malignancies are concerned, the problem is perhaps even more formi-tries. Current treatment approaches to most neoplasias are as crude as their targets are indomitable: patients dable. A related difficulty is that administration of gene vectors and adjuvant drugs may be imperfect. Where are carved with sharp instruments, poisoned, or subjected to dangerous radiation. This is not to deprecate solid tumors are concerned, penetration of gene vectors to all cells may be a serious impediment, because cells the dedication of oncologists, which often reaches heroic proportions; it is only to point out that our arma-embedded in large tumor masses are often difficult to access. The problem extends to drugs administered as mentarium is weak. Under these circumstances, one adjuvants for gene therapy. If these compounds are can easily understand why physicians, scientists, and unable to penetrate to the genetically engineered cells, patients look upon gene therapy with eager anticipathey will, of course, be ineffective. tion.
A potential approach to these obstacles is genetic The article by Kanai et al. in this issue of HEPATOLengineering of enhanced sensitivity to host immune OGY explores the use of adenovirus vectors expressing mechanisms. Golumbek et al. 2 expressed interleukinthe herpes thymidine kinase (tk) gene under the regu-4 in renal carcinoma (Renca) cells. Not only were these lation of the a-fetoprotein (AFP) promoter for condicells rejected, but the animals were rendered capable tional ablation of hepatoma cells. In this in vitro model of eliminating inocula of non-interleukin-4-producing of gene therapy for hepatocellular carcinoma (HCC),
Renca cells introduced at a distant site. Similarly, inthe authors report that the DNA construct is highly jection of interleukin-12-producing fibroblasts into effective in killing hepatoma cells that express AFP sarcomas of mice dramatically impaired tumor growth, when ganciclovir (GCV) is administered, and, that and systemic interleukin-12 administration reduced neighboring, nontransfected cells are efficiently killed established lung metastases. 3 These findings hold some by a bystander effect. The experimental design is based promise for immune enhancement as a gene therapy upon the fact that AFP, normally expressed only in the approach, though some serious problems may still refetus, is reactivated in many cases of HCC. Thus, the main (see below). experiments exploit unique patterns of gene expression It has also been proposed 4 that, under circumstances in malignant cells to target those cells specifically for where a substantial tumor risk is known to exist, ablachemotherapeutic ablation. The publication of these tion genes might be prophylactically inserted into the findings provides a timely opportunity to examine the organ(s) at risk. An example of this strategy would be problems posed by application of such gene therapy insertion of AFP/tk genes into livers of individuals with chronic active hepatitis. If a founder tumor cell possessed the ablation gene prior to neoplastic degenera-Abbreviations: tk, herpes thymidine kinase gene; AFP, a-fetoprotein; HCC, tion, all mitotic descendants of that founder, regardless hepatocellular carcinoma; GCV, ganciclovir.