Chapter 6: Specific Disease/Solid Tumors
Studies in solid tumors are technically different than studies in hematologic neoplasms because solid tumor most commonly are present as three-dimensional aggregates of cohesive cells, while hematologic neoplasms are almost exclusively discohesive. Studies by Teicher and Kerbel in murine tumors showed that in vitro drug activity correlated with in vivo drug activity when tumors were tested in vitro as three dimensional clusters, but not when they were tested in two dimensional monolayers . There is now an extensive literature on what has been labeled "multicellular resistance" [129-131].
All published clinical correlations with true fresh tumor assays with cell death endpoints have tested the tumor cells largely in the form of three dimensional clusters. The only exception to this statement is the non-small cell lung cancer study of the NCI-Navy medical oncology group, in which subcultured cells (not true fresh tumors) were tested in monolayer culture . This latter study not surprisingly showed poor correlations; all of the other cited studies, which used true fresh (non-subcultured) tumor cells tested largely as three dimensional cell clusters (and not in monolayer culture), showed good correlations.
The solid tumors which have received the greatest degree of study are gastrointestinal adenocarcinomas (colon and gastric adenocarcinoma), breast cancer, and ovarian cancer. There have been relatively few clinical correlations published in the cases of melanoma, soft tissue sarcoma, glioblastoma, and squamous cell carcinomas in general.
In the case of gastrointestinal neoplasms, there have been 129 published correlations between assay results and clinical response Table 1 (page 1) and Table 1 (page 2) . Overall, patients treated with drugs having good activity in the assays had a 48% response rate, while those treated with drugs having poor activity in the assays had a response rate of less than 1%, in a population of patients who overall had a 11% response rate. Also reported in many additional patients were positive associations between assay results and patient survival [64,133,134].
Colon and stomach cancer have been studied mainly with the MTT endpoint, most prominently by Kubota and colleagues at Keio University in Tokyo [64,133,135-140], but also by other Japanese investigators [63,134,141-144]. The Keio group has published studies with the MTT assay in which tumors were cultured both in suspension and also as macroclusters (0.5 mm tissue fragments) using an "in vivo-like" culture technology developed by Hoffman . These studies showed correlations between assay results and both response and survival [64,133,135-139]. These studies support the relevance of the MTT endpoint for fluoropyrimidine-based chemotherapy . As MTT measures mitochondrial function, the response and survival correlations raise the interesting possibility that fluoropyrimidine activity in colon cancer may be directed at (and mediated through) autonomously-replicating mitochondria. What is most important here, however, is that the MTT endpoint has been shown to identify colon and gastric cancer patients who will have relatively favorable versus relatively unfavorable outcomes when treated with fluoropyrimidine-based therapy. This is very important for the reason that the cell death endpoint had previously been most problematic for fluoropyrimidines, of the currently FDA-approved cytotoxic drugs.
Taken in the broad context of the entire literature, these studies provide important confirmation of the broad (with respect to both drugs and tumor types) applicability of cell death assays. The issue of whether the MTT assay or measurements of thymidylate synthetase  is more accurate in gauging probability of response to specific types of fluoropyrimidine-based therapy awaits future head to head comparisons.
In the case of ovarian cancer, there have been 328 published correlations between assay results and clinical response Table 1 (page 2) . Overall, patients treated with drugs having good activity in the assays had a 77% response rate, while those treated with drugs having poor activity in the assays had a response rate of 11%, in a population of patients who overall had a 51% response rate. Also reported were highly positive associations between assay results and patient survival [147,163].
Kurbacher and colleagues treated 25 patients with ovarian cancer with ATP-assay-directed chemotherapy and compared outcomes with 30 non-randomized but clinically well-matched controls . In the control group, there was a response rate of 37% (2 complete responders), with median progression-free survival of 20 weeks and median overall survival of 69 weeks. In the assay-directed group, there was a response rate of 64% (8 complete responders), with a median progression-free survival of 50 weeks (P2=0.003) and a median overall survival of 97 weeks (P2=0.145). Assay directed therapy also produced a greater benefit with respect to both response rate and progression-free survival in the subgroup of patients with platinum-resistant disease. A current multi-institutional, international trial is currently in progress to determine if assay-directed therapy is superior to empiric therapy.
In the case of breast cancer, there are a total of 179 published correlations between assay results and patient treatment Table 1 (page 1). Patients treated with assay "sensitive" drugs had an 82% response rate. Patients treated with assay "resistant" drugs had a 7.7% response rate. The overall response rate for the patients in the studies was 66%.
Xu and colleagues treated 73 breast cancer patients on the basis of MTT-assay directed chemotherapy, and compared outcomes with 73 patients treated with "physician's choice" chemotherapy . This was also a non-randomized study, but the patients receiving assay-directed therapy actually had less favorable prognostic factors, such as having significantly more sites of disease (the author informed me in a personal communication that the patients at her medical center with unfavorable disease were more often referred for biopsy and assay-directed therapy, while patients with more favorable disease were more likely to receive standard empiric chemotherapy). The response rate of the assay-directed group was 77%, while the response rate for the empiric therapy group was 44%. In a small group of 10 patients who received assays but in which no active drugs were identified, empiric therapy was given with no responses (0% response rate). One year survivorship for the two groups was 74% for assay-directed therapy and 67% for empiric therapy. Three year survivorships were 25% and 19%, respectively. Five year survivorships were 20.5% and 12.3%, respectively.
The above study showed a clear response advantage to assay-directed therapy and a trend for a survival advantage, despite less favorable prognostic factors for the group receiving assay-directed therapy. The lack of statistical significance for survival is no doubt owing to the small numbers of patients enrolled in the study. Putting things into perspective, the adjuvant Cancer and Acute Leukemia Group B study comparing doxorubicin/cyclophosphamide with and without Taxol required 2,000 patients to show an absolute 2% difference in survival. And yet triple drug therapy has now become the standard of care in this setting. It also required a meta-analysis of studies totalling close to 50,000 patients to establish a small survival advantage for adjuvant chemotherapy of post-menopausal patients.
Return to HTAJ Home Page
Return to Title/Abstract/Chapter Table of Contents
Return to Chapter 5: Hematologic Neoplasms
Proceed to Chapter 7: Editorial Conclusions