Background: Many studies have shown strong correlations between assay results, chemotherapy response, and patient survival in ovarian cancer and other neoplasms. Such correlations are the traditional criteria used to evaluate all laboratory and radiographic tests by the FDA, other regulatory agencies, and the medical profession in general. The following studies are, in fact, unprecedented attempts to determine the extent to which treatment outcomes may be improved through the use of laboratory tests.
Kurbacher and colleagues treated 25 previously-treated patients with ovarian cancer (in 1st through 5th relapse) with assay-directed chemotherapy and compared outcomes with 30 non-randomized but clinically well-matched controls (Kurbacher, CM, et al. Anticancer Drugs 9:51-57,'98). In the control group, there was a response rate of 37% (2 complete responders), with median progression-free survival of 5 months and median overall survival of 17 months. In the assay-directed group, there was a response rate of 64% (8 complete responders), with a median progression-free survival of 12.5 months (P2=0.003) and a median overall survival of 24 months (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. In an update presented at the American Society of Clinical Oncology (ASCO) meeting in 2001, the authors reported treating 59 previously-treated patients, including 28 with initially platinum-resistant disease. Median survival for patients with platinum-sensitive disease (with 2nd through 5th line therapy) was 24 months and for platinum-resistant disease was 21 months. A current multi-institutional, international trial is currently in progress to further determine whether assay-directed therapy is superior to empiric therapy.
Loizzi and colleagues treated 50 ovarian cancer patients in their first recurrence following platinum-based therapy with assay-directed therapy and compared clinical outcomes with 50 additional non-randomized but well-matched patients who received chemotherapy without assay information (results presented at the 34th Annual Meeting of the Society of Gynecologic Oncologists New Orleans, 2003, Abstract 189). There were differences in the impact of assays between patients who had primary platinum-sensitive disease versus platinum-resistant disease (unlike the Kurbacher study, in which there were equal benefits to patients with platinum-sensitive and platinum-resistant disease). In the group with initial platinum-sensitive disease, response rates were 65% for assay-directed chemotherapy versus 35% for empiric chemotherapy (P=0.02). Overall and progression-free median survival times were 38 and 15 months, respectively, in the assay-directed group versus 21 and 7 months for empirically-treated control patients (P=0.005, P=0.0002). However, in the platinum-resistant group, there was no improvement for assay-directed therapy. Response rates were 21% versus 16% (not significant). Overall median survivals were 13 months versus 12 months (n.s.). Median progression-free survivals were 5 months versus 6 months (n.s.). (n.b. this study has now been published in full: Loizzi, V, et al. Am J Obstet Gynecol. 2003 Nov;189(5):1301-7).
Weisenthal, et al recently completed an analysis of the overall survival of 549 ovarian cancer patients with tumors submitted to our laboratory for testing between January, 1993 and January, 2001. Kaplan-Meier survival curve analysis shows median survivals of 44 months for 1st line chemotherapy of previously-untreated patients, 41 months for 2nd line chemotherapy of patients meeting the clinical definition of initially platinum sensitive-disease (platinum-free interval > 6 months), 27 months for platinum-resistant disease (platinum-free interval 6 months or less), 23 months for patients receiving 3rd, 4th, or 5th line therapy (chemotherapy-free interval > 6 months), and 7 months for patients receiving 3rd, 4th, or 5th line therapy (chemotherapy-free interval 6 months or less). Click here to view Survival Curves.
We have also recently performed an analysis of the long-term survival of ovarian cancer patients who had tumor biopsy specimens referred to our laboratory for cell culture drug resistance testing (CCDRT or "chemosensitivity testing"), with the data analyzed as a function of whether or not the attempted CCDRT was evaluable or inevaluable (note that approximately 95% of specimens submitted for testing do yield an evaluable result, while approximately 5% are inevaluable for reasons relating to poor viability of the specimen, insufficient yield of tumor cells in the specimen, or other reasons). Comparing the survival of patients with evaluable assays (which resulted in CCDRT data to guide therapy) versus inevaluable assays (in which CCDRT data to guide therapy were not available), there were significant differences favoring patients in whom CCDRT was evaluable and in which results were reported to the referring oncologists. Click here to review these studies.
In considering the results of our own analysis (Weisenthal, et al, above), it is important to note that these survival outcomes were not obtained in the context of a formal clinical trial. Rather, these data represent the survival of patients for whom tumor biopsy specimens were submitted to our laboratory for cell culture drug resistance testing (CCDRT). We do not have any information regarding what forms of chemotherapy were actually administered to these patients. Since all of these specimens were submitted on a non-investigational basis and since patients (or their insurance companies) were charged an average of nearly $2,000 for the performance of the tests, one presumes that the results of the testing were considered in treatment planning, but it would not be surprising if all patients were not treated strictly according to the assay results, particularly in the cases of first line chemotherapy. This may or may not explain why the survival of both platinum-sensitive and resistant-patients receiving 2nd line chemotherapy were not significantly different than the survival of previously-untreated patients (i.e. if "2nd line" patients were treated more strictly according to assay results than were "1st line" patients, then this may have provided the "2nd line" patients with disproportional benefit, compared to the "1st line" patients).
One additional point to consider is the superior survival of the platinum-resistant patients in the Kurbacher study and in the Weisenthal series, compared to the Loizzi study. It is noteworthy that both Kurbacher and Weisenthal employed assays using cell death endpoints, while the Loizzi study employed a cell proliferation endpoint. As discussed elsewhere on this website, I believe (based on personal experience and on the basis of published data) that cell death assays may be more robust with respect to indentifying clinically-active drugs than are cell proliferation assays. In 1991-2, I proposed a clinical trial to make a head to head comparison between the cell death and cell proliferation endpoints in ovarian cancer to the Gynecologic Oncology Group (GOG). Despite great efforts (including the writing of a draft protocol), the GOG declined to do the study, in favor of performing dozens of studies of empiric chemotherapy which have not identified better chemotherapy treatment regimens in the ensuing 12 years.
It is time to abandon the empiric, "one sized fits all" treatment paradigm in favor of recognizing what has been known for more than 40 years, namely that ovarian and other forms of cancer represent heterogenous diseases, where the tumors of different patients have different responses to chemotherapy, and where important treatment advances will require individualizing treatment based on testing the individual properties of each patient's cancer.