Special Feature
Highlights from SGO 2008: Ovarian Cancer
By Robert L. Coleman, MD, Associate Professor, University of Texas; M.D. Anderson Cancer Center, Houston, is Associate Editor for OB/GYN Clinical Alert
The national organization in the United States dedicated specifically to the care of women with gynecological malignancy is the Society of Gynecologic Oncologists. The organization meets once annually to present and discuss some of the most innovative, provocative and relevant investigational research and practices across the spectrum of malignancies that uniquely affect women. The conference continues to grow, with a large international presence, and as such, is challenged in its ability to disseminate all the information presented to its entire constituency. In this Special Feature, I will summarize a few of the many notable projects discussed at the conference and will specifically focus on ovarian cancer in order to explore, in greater depth, research in the many important areas of clinical interest. The abstracts can be found online: http://www.sciencedirect.com.
Ovarian Cancer Screening/Early Detection
It may be surprising to note that ovarian cancer, stage for stage, shares similar survival to other solid tumors. The problem is more than three quarters of patients are diagnosed when the disease is disseminated. It is, therefore, not surprising that efforts to "stage migrate" the population of incident ovarian cancer to more limited disease is a worldwide endeavor. Over the past year, an emphasis has been placed on alerting women that the symptoms associated with ovarian cancer at diagnosis (symptom index) can be distinguished from day-to-day maladies by their constellation, severity and duration.1 Expanding on these observations, Andersen and colleagues at the University of Washington evaluated whether the addition of the symptom index (SI) could improve the sensitivity and specificity of an experimental CA125-based early detection program. A prospective case control study of women at high risk (based on family history) for ovarian cancer was conducted. Characteristics of abnormal baseline testing included: symptoms that occurred > 12 times/month, documented in the previous year (positive SI), and a CA125 level ≥30 U/mL. A total of 254 women were screened. One hundred thirty-one were diagnosed with benign masses and 75 were diagnosed with ovarian cancer (31 early, 41 late stage). The majority of the cancer cases (n=40, 53%) had both a positive CA125 and a positive SI. However, 8 (11%) of the cancer cases had a negative CA125 and a positive SI — similar to the 8 (11%) cancer cases with a negative CA125 and a negative SI. Nineteen (25%) had a positive CA125 without symptoms. Logistical regression analysis demonstrated that the SI independently predicted ovarian cancer (OR=11.5, 95% CI 4.6-28.7). Both modalities (CA125 and SI) had low sensitivity (<80%), which improved when combined (sensitivity overall 89.3%). In all, half of the CA125 negative tumors were picked up by symptoms. However, gains in sensitivity were made at the expense of specificity (84%), meaning more abnormal testing occurred in patients without cancer. This has been a major problem with population-based screening, particularly among low prevalent cancer. In addition, the composite marker (CA125 and SI) performed little better than CA125 alone in early stage disease. Nonetheless patient and physician education regarding these symptoms and their potential link to early diagnosis appears warranted while further investigation is undertaken.
More traditional screening efforts in ovarian cancer have centered on serum and urine biomarkers. Here, identification of differentially expressed proteins accompanying early dysregulation of epithelial ovarian cell growth is the focus.2-4 Four reports using biomarkers to either distinguish (cancer or benign) patients with abdominal-pelvic masses or to detect stage I disease were presented. Each of these studies used new investigational biomarkers as additions to CA125. The collective experience highlights modest gains to the (under) performance of CA125 alone. However, the difficulty with each of these studies is their conduct in highly prevalent situations (postmenopausal women, women with pelvic masses, highly prevalent stage I disease, etc.). Ultimately, these types of studies are important to develop promising "cocktails" for future investigation but raise concerns predicated on the characteristics of a successful screening tool: affordable, accessible, tolerable, reproducible and identifies preclinical or early disease.
Ovarian Cancer Therapeutics:
First-Line Therapy
The high morbidity and mortality of recurrent ovarian cancer makes the development of new and effective primary strategies a major priority. This year's meeting included several presentations of new potential therapeutic avenues, as well as, alterations to highly toxic but effective intraperitoneal (IP) chemotherapy.5 In regard to the latter, a survey addressed to the SGO membership queried how clinicians dealt with the paradox. Naumann et al, reported that while the vast majority of gynecologic oncologists offered IP therapy to their patients, 58% modified the regimen to make it more tolerable and outpatient. There was wide variation in how this was accomplished with no more than 18% of respondents doing the same type of modification. Further, there was use of IP therapy outside of its published investigational population cohort, including 1 in 8 clinicians using IP therapy in suboptimal patients and nearly one-third using IP "sometimes" in early stage ovarian cancer. The survey clearly demonstrates the need for continued study in order to optimize IP infusion schedules, agents and doses.
HMG-CoA reductase inhibitors or "statins" commonly used to treat dyslipidemia may be relevant in cancer therapy. These agents have been documented in preclinical investigation affecting tumor biology by promoting tumor apoptosis and preventing metastases. In several solid tumors (breast, prostate, etc), the use of these agents has been associated with favorable (and unfavorable) effects on survival.6 Their use in ovarian cancer has not been widely reported. To address this hypothesis, Elmore and colleagues conducted a retrospective review of statin use in ovarian cancer patients undergoing primary cytoreductive surgery. Users (n=17) were compared with non-users (n=109) for effects on progression-free and overall survival. Despite being older and more commonly having the diagnosis of diabetes mellitus, users were found to have superior PFS and OS — both significantly. Even after controlling for known prognostic factors, there was a 55% reduction in the hazard for death in statin users. Although this study is a small, single institutional, retrospective analysis, statin use at the time of initial cytoreductive surgery has hypothetical merit. Given the variable observations of statin use on significant outcomes in other diseases and the possibility that statin use may be a surrogate marker for a more direct and relevant prognostic factor, the impact of statin use on ovarian cancer patients warrants further careful investigation.
A perennial favorite at the annual meeting is a discussion of how surgery influences outcome. Over the years, several different approaches and questions have been addressed in generally retrospective or cohort studies given the difficulty of performing a primary randomized control trial. Although not a new topic, neoadjuvant chemotherapy ahead of surgery appears to be gaining some momentum in primary management.7 The principle driving mechanism is the ability to more easily (and safely) perform tumor debulking, achieving high rates of minimal volume residual. In this light and supported by the positive data of IP therapy in minimal volume residual patients, there has been some interest in combining these two approaches in a study cohort — that is sequencing therapy as: neoadjuvant intravenous chemotherapy, followed by interval tumor debulking, followed by IP therapy after surgery. Tiersten and colleagues performed the strategy in 56 patients with advanced bulky stage III and IV disease. Standard intravenous chemotherapy was administered and in those with CA125 reduction of more than 50%, surgery was performed. Overall, 78% of women initiated on treatment finished 3 cycles, and of these 80% underwent surgery. Three quarters of the surgical population received post-operative chemotherapy. Overall survival was similar to other reports of suboptimal patients. The dramatic dropout of patients at each treatment node highlights the difficulty in studying poor performance status patients. Although the strategy is appealing and takes advantage of the merits of both neoadjuvant therapy and IP chemotherapy, it will need prospective evaluation in less infirmed patients to evaluate its merit in our primary armamentarium.
Ovarian Cancer Therapeutics:
Recurrent Disease
Recurrent ovarian cancer remains one of the most difficult and trying objectives of therapy given the almost universal mortality of patients so diagnosed, the impact of mounting, cumulative toxicities from treatment and the need to enable, as much as possible, quality of life. Because of the insufficient level of clinical activity, patients with recurrent disease are good candidates for novel therapeutics investigation. The pre-clinical and clinical experience of several new agents was presented at the annual meeting.
Without a doubt, the most visible strategy for developmental therapeutics in ovarian cancer is anti-angiogenesis.8 Currently, multiple agents using direct and indirect approaches to blocking the formation of new vasculature in tumors are entering the clinical domain. The most mature in ovarian cancer is the vascular endothelial growth factor (VEGF) ligand-binding antibody, bevacizumab. This agent is in phase III investigation in primary and recurrent ovarian cancer and is being studied around the world in different cooperative groups. Richardson and colleagues examined the efficacy of a triplet (gemcitabine, carboplatin or cisplatin, and bevacizumab) in a single-institution study.
Thirty-five patients were retrospectively analyzed. A median 6 cycles was administered to these patients with recurrent disease. Toxicity was consistent with each of these agents including 2 patients with bowel perforations and one enterocutaneous fistula. However, response to treatment was impressive with nearly half achieving complete remission. Joining this report was a study, combining two biological agents for patients with recurrent disease. The second therapy, sorafenib, targets several mechanisms supporting angiogenesis and cellular signaling. The two are synergistic in preclinical models of ovarian cancer and work through complementary targets but heretofore, are untested in ovarian cancer patients. Clinical tolerance was the primary objectives of this phase I dose-escalating study. Patients were heavily pretreated (mean 6 prior therapies). Dose reductions were frequently necessary among the 16 enrolled patients. The final recommended dose for phase II investigation was at the lower levels of documented efficacy for each agent, when administered as single agents. Efficacy, measured by response and not a primary endpoint, was seen in 7 of 15 evaluable patients. While toxicity observations were important, especially non-hematological events such as hand-foot syndrome and proteinuria, the additional information regarding the lack of CA125 association to responses, reminds us that traditional biomarkers may be unreliable in assessing biologic therapy. Further work with this combination is warranted.
Platinum doublets (taxane/platinum and gemcitabine/carboplatinum) compared to single agent platinum improve PFS and OS in women with epithelial ovarian cancer. A novel antifolate, pemetrexed, inhibits multiple enzymes in the DNA synthesis pathway, and synergy between platinum and pemetrexed is documented. Matulonis et al, conducted a phase II trial to evaluate the impact of carboplatin and pemetrexed on platinum-sensitive recurrent ovarian cancer with a primary endpoint of response. The combination was administered on a 3-week cycle for 6 cycles. Of the 44 patients enrolled, 23 (52%) patients had a confirmed partial response. Clinical benefit, measured as objective response plus stable disease was observed in nearly 90% of the patients. Grade 3 and 4 hematologic toxicity was uncommon but included neutropenia, thrombocytopenia and anemia. The most common grade 3 non-hematologic toxicity carboplatin reaction occurred in just over one-third. Median PFS was 7.6 months (95% CI, 6.4-10.2) months, and mean OS was 20.3 months (median OS has not yet been reached). The authors concluded that carboplatin and pemetrexed is an active regimen for women with platinum-sensitive disease. In addition, the toxicities were manageable and predictable, and PFS and OS were comparable to other platinum doublets. Given these observations, the combination appears reasonable for further clinical development.
Conclusions
The research agenda in ovarian cancer is diverse and complex and many facets of this disease were presented/discussed at the annual meeting. Intense interest surrounds this primary, as the topic was a focus for a substantial proportion of the meeting as well as the numerous breakout and educational sessions. Although, many advances are being made, the necessary resources (patients and funding) to fully investigate the varied questions are limited, which will necessarily cause the research community to develop, in addition to the novel treatment, novel biomarkers and statistical methodology. Fortunately, these paradigms are already being explored.
References
- Goff BA, et al. Development of an ovarian cancer symptom index: possibilities for earlier detection. Cancer. 2007;109:221-227.
- Bast RC Jr., et al. New tumor markers: CA125 and beyond. Int J Gynecol Cancer. 15 Suppl. 2007;3:274-281.
- Grisaru D, et al. Microarray expression identification of differentially expressed genes in serous epithelial ovarian cancer compared with bulk normal ovarian tissue and ovarian surface scrapings. Oncol Rep. 2007;18:1347-1356.
- Psyrri A, et al. Human tissue kallikrein 7, a novel biomarker for advanced ovarian carcinoma using a novel in situ quantitative method of protein expression. Ann Oncol. 2008;Mar 5 [Epub ahead of print]
- Armstrong DK, et al. Intraperitoneal cisplatin and paclitaxel in ovarian cancer. N Engl J Med. 2006;354:34-43.
- Taylor ML, et al. Statins and cancer: a meta-analysis of case-control studies. Eur J Cancer Prev. 2008;17:259-268.
- Bristow RE, Chi DS. Platinum-based neoadjuvant chemotherapy and interval surgical cytoreduction for advanced ovarian cancer: A meta-analysis. Gynecol Oncol. 2006;103:1070-1076.
- Spannuth WA, et al. Angiogenesis as a strategic target for ovarian cancer therapy. Nat Clin Pract Oncol. 2008;5:194-204.
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