CA125 to Monitor for Relapse in Epithelial Ovarian Cancer

Apr 23, 2010


Jonathan S. Berek, MD, MMS
Professor and Chair, Department of Obstetrics & Gynecology, Stanford University School of Medicine, Director, Women's Cancer Program, Division of Gynecologic Oncology, Stanford Cancer Center

CA125 measurements are very useful blood tests for monitoring response to the initial therapy for the majority of patients with epithelial ovarian cancer. However, the value of CA125 levels to follow patients who are in clinical remission after first-line chemotherapy recently has been challenged in a study conducted by Gordon J. S. Rustin, MD, MBBS, and colleagues presented at the 2009 ASCO Annual Meeting. It is well known that in most patients the CA125 levels tends on average to rise several months earlier than the recognition of physical or radiologic signs of recurrence, but the effect on survival of the timing of re-initiation of chemotherapy has been unclear. In Dr. Rustin's trial, patients were randomly selected to have CA125 levels monitored for biochemical relapse versus using other clinical signs of recurrence, and each of these signs were then used to initiate second-line chemotherapy. On average, the CA125 levels rose 5 to 6 months prior to other signs of disease recurrence. The authors reported that the overall survival of patients was not improved by initiating earlier second-line chemotherapy based only on the rising CA125 levels. They recommended that the routine practice of monitoring of CA125 levels in these women with ovarian cancer in clinical remission could be modified—a controversial approach. It is unclear how this study will be applied clinically, and Stanley B. Kaye, MD, and Martin E. Gore, MD, PhD, discuss the issues that affect these important therapeutic decisions in ovarian cancer.

Should Routine CA125 Screening Be Discontinued?

Stanley B. Kaye, MD, FRCP, FRCR
Head of the Section of Medicine, The Institute of Cancer Research, Royal Marsden Hospital; London, United Kingdom

The Medical Research Council (MRC)/European Organisation for the Research and Treatment of Cancer (EORTC) OVO5/5595 study was a well-conducted randomized trial that provided a clear conclusion to the question addressed: does early re-treatment of asymptomatic recurrent ovarian cancer based on CA125 measurement carry any survival advantage? The answer is clearly no, and many clinicians, particularly in Europe, will have concluded that their practice in delaying treatment has been justified. This study may well have a significant impact in those units where early treatment has been the norm, but should the data presented also lead to discontinuation of routine CA125 measurement? There, many clinicians may hesitate to change practice.

The argument against discontinuation is that this excellent clinical trial took place during a 10-year period (1996-2006) when the options for treatment were relatively limited. Although the main planks of chemotherapy remain unchanged (taxanes and platinum compounds), the treatment of ovarian cancer is beginning to move forward in increasingly exciting ways. The era of targeted therapy has arrived; novel agents, such as those targeting angiogenesis, or those focusing on homologous recombination (HR) deficient cells, are beginning to have an effect.1 While randomized trial data are still awaited, our hope is that the paradigms for ovarian cancer treatment will change. With that change, it is quite possible that earlier recognition of disease relapse will become increasingly important. For that reason, a “tablets of stone” policy of discontinuing
CA125 measurement routinely may be a mistake.

Of course, a policy of continuation of CA125 monitoring invites the major problem of patient “addiction” to CA125 results.2 The anxiety that surrounds clinic visits, and the obsessive recording of CA125 data, is well known to clinicians. This has to be weighed against the potential benefits, and this trial clearly has to be considered in that equation.

Some clinicians will say that another reason that early detection of relapse is important is to allow a surgically resectable recurrence to be detected and removed. This trial does not directly address that issue and, while ongoing trials are addressing the role of surgery for disease relapse in a randomized fashion, many clinicians will continue to adopt this approach, encouraged by the experience of clinical benefit in their own practices.

Finally, there may be merit in looking in more detail at the rate of rise of CA125 in patients with asymptomatic first recurrence of ovarian cancer. Data from our unit indicate that patients with a CA125 doubling-time of less than 3 weeks have a median survival of 9.5 months, compared to those with a doubling-time of more than 11 weeks, who have a median survival of 34.5 months.3 As new treatments evolve, this information may well be one important factor in decision-making regarding the optimal time to begin relapse therapy.

To conclude, Dr. Rustin and all the trialists are to be congratulated for a thought-provoking and potentially practice-changing trial. As the treatment of ovarian cancer enters its most exciting phase for many years, the trial provides an important framework for clinicians to use, for the benefit of all their patients with ovarian cancer. Dr. Rustin and colleagues advocate that patients are offered an informed choice regarding CA125 monitoring, and clearly an individualized approach has merit.4 However, at this particular time, routine CA125 monitoring of patients following initial treatment continues to be a justified procedure.


  1. Yap T, Carden CP, Kaye SB. Nat Rev Cancer. 2009;9:167-181.
  2. Tummala MK, McGuire WP. J Clin Oncol. 2007;25:3570-3571.
  3. Karavasilis V, Thomas K, Harrison M, et al. J Clin Oncol. 2008;26 (May 20 suppl, abstr 5544).
  4. Hopkins ML, Coyle D, Le T, et al. Curr Oncol. 2007;14:167-172.

The Timing of Treatment for Relapsed Disease

Martin E. Gore, MD, PhD, FRCP
Professor of Cancer Medicine, Royal Marsden Hospital, Institute of Cancer Research; London, United Kingdom

The MRC and EORTC are to be congratulated for performing the OV05/5955 study, which is a landmark trial in challenging some firmly held views on how best to treat patients with advanced ovarian cancer that has relapsed following first-line chemotherapy. We all have genuine admiration for the dogged determination that was required to complete this work.

Many groups in Europe, and perhaps especially in the United Kingdom, have for some time questioned the practice of starting chemotherapy for relapse on the basis of a rising CA125 alone.1

OV05/5955 demonstrates that there is no survival advantage to treating patients on a rising CA125 compared to waiting until the patient develops symptoms. OV05/5955 does not inform us as to the optimum time to commence treatment for relapse because some unaddressed questions remain. There is evidence to show that one can leave a patient too long before treating a relapse, i.e., waiting until the disease gets bulky can reduce the chance of a response.2 In addition, many patients who develop symptoms at relapse have evidence of sub-acute obstruction. There are conflicting data on the significance of intestinal obstruction on the chemo-responsiveness of relapsed disease3,4 but most feel that patients with conventionally defined resistant relapse are unlikely to benefit from chemotherapy. Leaving patients to become obstructed before commencing chemotherapy may reduce their chance of a response and therefore possibly impact survival. It would be important to know the number of patients with disease above 5 cm and/or in any degree of obstruction at the time of the commencement of second-line therapy and how these two parameters relate to patients’ platinum-free interval.

There is now firm evidence that the choice of treatment at relapse can have an effect on overall survival and we need to be aware of the advances that have been made in this area since the commencement of this study.5-7 In OV05/5955 the median time to second-line therapy for the patients who were treated “early” was 0.8 months, whereas it was 5.6 months for those whose treatment was delayed—yet the treatments in the two arms appear to be identical. Oncologists would usually tailor treatment at relapse according to whether or not the patient had “sensitive” or “insensitive” disease, based on the platinum-free interval and the recognised cusp for presumed continued platinum sensitivity is 6 months.8,9 One might expect patients in the early-treatment group to have a lower rate of platinum-based therapy than those whose treatment was delayed. Furthermore, from data that has emerged in the last few years, we know there is an advantage to using combination platinum-based therapy over single agent platinum for patients who relapse with a platinum-free interval of over 6 months.10,11 The fact the two randomized groups in this trial are well-balanced for the second-line therapies received is a worry, not a comfort.

Finally, there are emerging data on the role of surgery in patients who experience disease recurrence after first-line chemotherapy for ovarian cancer12 and, as with the questions set out above, the data from OV05/5955 cannot be taken in isolation. We need to place them into the context of our knowledge of the natural history and biology of relapsed ovarian cancer and the predictors of benefit from both surgical and medical interventions.

The stress of follow-up on our patients and the development of CA125 “addiction” is an issue of huge importance.13 The MRC/EORTC OV05/5955 study rightly re-focuses our attention on this phenomenon. Particular congratulations are due to Dr. Rustin, who has driven forward a number of concepts involving the utility of serum CA125 measurements, including this latest important study.


  1. Gore ME. Controversies in Surveillance Options for Patients after Initial Treatment for Ovarian Cancer. 2006 ASCO Educational Book.
  2. Eisenhauer EA, Vermorken JB, van Glabbeke M. Ann Oncol. 1997;8:963-968.
  3. Bryan DN, Radbod R, Berek JS. Int J Gynecol Cancer. 2006;16:125-134.
  4. Abu-Rustum NR, Barakat RR, Venkatraman E, et al. Gynecol Oncol. 1997;64:493-495.
  5. Gore M, Oza A, Rustin G, et al. Eur J Cancer. 2002;38:57-63.
  6. Parmar MK, Ledermann JA, Colombo N, et al. Lancet. 2003;361:2099-2106.
  7. Meier W, du Bois A, Reuss A, et al. Gynecol Oncol. 2009;114:199-205.
  8. Gore ME, Fryatt I, Wiltshaw E, et al. Gynecol Oncol. 1990;36:207-211.
  9. Markman M, Rothman R, Hakes T, et al. J Clin Oncol. 1991;9:389-393.
  10. Pfisterer J, Plante M, Vergote I, et al. J Clin Oncol. 2006;24:4699-4707.
  11. Alberts DS, Liu PY, Wilczynski SP, et al. Gynecol Oncol. 2008;108:90-94.
  12. Harter P, du Bois A, Hahmann M, et al. Ann Surg Oncol. 2006;13:1702-1710.
  13. Harries M, Gore M. Lancet Oncol. 2002;3:537-545.

Current Controversies in Oncology is a forum for the exchange of views on topical issues in the field of oncology. The views and opinions expressed therein are those of the authors alone. They do not necessarily reflect the views or positions of the Editor or of the American Society of Clinical Oncology.

Back to Top