Annals of Oncology 22 (Supplement 8): viii45–viii48, 2011

doi:10.1093/annonc/mdr471

symposium article

Follow-up with CA125 after primary therapy of advancedovarian cancer has major implications for treatmentoutcome and trial performances and should not beroutinely performed

G. J. S. Rustin*Department of Medical Oncology, Mount Vernon Cancer Centre, Northwood, UK

Background: Only one randomized trial has examined the value of performing routine CA125 measurements during

follow-up of ovarian cancer. The results of this trial and implications of frequent CA125 measurements are examined.

Patients and methods: The Medical Research Council OV05/European Organisation for Research and Treatment

of Cancer 55955 trial enrolled 1442 patients with a CA125 level within the normal range following platinum-based

chemotherapy for epithelial ovarian cancer. If CA125 levels rose to more than twice the upper limit of normal, patients

were randomized to immediate or delayed chemotherapy.

Results: Those randomized in the early arm started chemotherapy a median of 4.8 months earlier than those on the

delayed arm. There was no difference in survival between the early and delayed arms.

Conclusions: Women should be advised not to have routine CA125 measurements, providing they are well and

have no symptoms suggesting relapse. In asymptomatic patients with a rising CA125 level, chemotherapy can be

delayed. Earlier stopping of maintenance therapy just because of rising CA125 might deny patients continuing benefit

from that therapy. Use of CA125 to define progression could result in platinum-sensitive patients being falsely

classified as platinum resistant.

Key words: CA125, ovarian cancer, follow-up, tumour progression, relapse

introduction

It has become standard practice around the world to performroutine CA125 measurements on women who have completedchemotherapy for ovarian cancer. The main reason for follow-up given by German ovarian cancer patients was that theythought it would lead to them living longer [1]. It is a poorreflection on the medical community that patients have beenled to believe an important concept for which there is noevidence. It is obvious that if earlier detection of relapse leads toimproved outcome, then the earlier the relapse is detected thebetter. However, if earlier detection does not improve outcome,an earlier diagnosis of relapse leads to patients living longerwith the knowledge that their cancer is relapsing yet deriving nobenefit from this knowledge.This article explains the results and implications of the only

randomized trial that has compared early versus delayedtreatment for relapsed ovarian cancer, providing for the firsttime the evidence that will allow follow-up policies to beappropriately developed. What is perhaps not appreciated is

how different follow-up policies can influence not only thetiming of treatment for relapse but also the duration ofmaintenance therapy. Different follow-up schedules can alsoinfluence the date of progression, which is a major endpoint inclinical trials. I will show how, using the latest research results,a sensible follow-up schedule can be applied [2].

the MRC OV05/EORTC 55955 trial

A woman with a rising CA125 concentration, who remainswell, without symptoms or signs of recurrent disease, presentsa major management dilemma. To resolve this dilemma, a trialwas designed to compare early treatment of relapse based ona rising CA125 level with delaying chemotherapy until signs orsymptoms of relapse [3]. The Medical Research Council (MRC)OV05/European Organisation for Research and Treatment ofCancer (EORTC) 55955 trial enrolled 1442 patients who hada complete remission with a CA125 level within the normalrange following platinum-based chemotherapy for epithelialovarian cancer. They had follow-up consultations and bloodtaken for CA125 measurement every 3 months. The results wereblinded from both the clinicians and patients, and were sent tothe MRC or EORTC trials centres. If the levels rose to more

sym

po

siu

ma

rtic

le

*Correspondence to : Dr G. Rustin, Department of Medical Oncology, Mount Vernon

Cancer Centre, Northwood, Middlesex HA6 2RN, UK. Tel: +44-1923-844190;

Fax: +44-1923-844840; E-mail: grustin@nhs.net

ª The Author 2011. Published by Oxford University Press on behalf of the European Society for Medical Oncology.

All rights reserved. For permissions, please email: journals.permissions@oup.com

by guest on January 3, 2014http://annonc.oxfordjournals.org/

Dow

nloaded from

than twice the upper limit of normal, patients wererandomized. In the ‘early’ arm, patients were told of the risinglevels and if confirmed they were asked to start relapsechemotherapy within 4 weeks. In the ‘delayed’ arm, patientscontinued on follow-up having blinded CA125 measurementsand only started chemotherapy if there were signs or symptomsof relapse.A total of 529 (37% of registered patients) were randomized,

whilst 421 (29%) were not randomized as they had not hada rise in CA125 and had no evidence of relapse. Nineteen percent of patients were not randomized because they relapsedclinically either without or at the same time as CA125 levelsrose, and a further 15% were not randomized for a variety ofreasons. Those randomized in the early arm startedchemotherapy a median of 4.8 months earlier than those in thedelayed arm, which confirms that CA125 can reliably detectasymptomatic relapsing ovarian cancer. Despite this earliertreatment, there was no difference in survival between the earlyand delayed arms {early arm: median, 25.7 months [95%confidence interval (CI), 23.0–27.9]; delayed arm: median, 27.1months (95% CI, 22.8–30.9); hazard ratio (HR): median, 0.98(95% CI, 0.80–1.20); P = 0.85} (Figure 1). Furthermore, third-line chemotherapy was started in those in the immediate arm4.6 months earlier than in those in the delayed arm [HR:median, 0.69 (95% CI, 0.58–0.83); P = 0.0001]. Although datawere not collected for therapy after third-line chemotherapy, itis likely that later lines of chemotherapy were also started earlierin those in the early arm. As those in the early arm had similarsurvival to those in the delayed arm, they must in total havereceived more lines of chemotherapy. Not surprisingly, therewas an earlier deterioration in the quality of life of thosepatients in the early arm.

implications for clinical practice

For the first time, women can be given evidence to help themdecide how they wish to be followed up after receiving first-linechemotherapy. They can decide:

� Not to have routine CA125 measurements providing they arewell and have no symptoms suggesting relapse.

� To continue having CA125 measurements, but not be told theresults. This option is particularly useful if they are ina clinical trial where routine CA125 measurements aremandated, but reduces anxiety over waiting for the result.

� To have routine CA125 measurements, so that they havemore warning as to when they might require relapsechemotherapy.

In patients whose CA125 is rising but who remainasymptomatic, the OV05/55955 trial clearly shows there is noneed to restart chemotherapy until symptoms develop.A computed tomography (CT) scan that shows minimal or nodisease can help patients to delay chemotherapy.It is only possible for patients to make sensible choices

regarding follow-up if they are given adequate information. It isadvisable to routinely discuss the rationale of follow-up withpatients when they complete first-line chemotherapy foradvanced ovarian cancer. If they are in complete remission witha normal CA125 level (the entry criteria for OV05/55955) theycan be told that they have almost a 50% chance of surviving 6years, from completion of their first-line chemotherapy. Theyneed to know that there is a high chance of relapse, but alsoa good chance of benefit from further chemotherapy. Myperception from debating this topic around the world is thatthose patients who are told by their doctors that relapseddisease can be treated but not cured are more likely to acceptthe results of OV05/55955 and not have routine CA125measurements performed. Those doctors who are happy thattheir patients believe that cure of relapsed ovarian cancer is stillpossible find it difficult to persuade their patients not to haveroutine CA125 measurements. Obviously, if there wasa reasonable chance of cure then we would all wish to treat asearly as possible.There are further requirements to enable safe follow-up of

ovarian cancer patients who opt for no routine CA 125measurements. They need to be told what symptomsshould prompt an early clinic appointment. A list of likelysymptoms can be supplied. It is essential that there arefacilities for patients to make urgent appointments. Easyavailability of a support nurse for telephone advice can bea great help.Some surgeons remain convinced that they can improve

the survival of selected patients with relapsed ovarian cancerand that the earlier the relapse is detected the better thechance of optimal surgery. A recent small study showed thatthose patients whose surgery was prompted by a rising CA125level had a worse survival than those whose relapse wasdetected by symptoms or routine examinations and scans [4].Hopefully, the DESKTOP 3 and GOG 213 trials will providerandomized trial data to demonstrate whether there isa survival benefit from surgery for relapse. Surgery is only ofvalue if it can result in complete macroscopic removal ofrecurrent cancer. There is therefore no point in doing routineCA125 measurements in those patients who are notcandidates for surgery, which is most patients during the first12 months following first-line therapy. Data from DESKTOP1 and 2 suggest this also includes all patients who had residualdisease after initial surgery, performance score >0 andpresence of ascites [5].

Figure 1. Kaplan–Meier plots for overall survival from randomization in

the OV05/55955 trial, reprinted with permission Lancet 2010; 376: 1155–

1163.

symposium article Annals of Oncology

viii46 | Rustin Volume 22 | Supplement 8 |December 2011

by guest on January 3, 2014http://annonc.oxfordjournals.org/

Dow

nloaded from

implications for research

As mentioned above, it is still reasonable to do routine CA125measurements as part of clinical trials, but patients should beoffered the chance of not having their results divulged if theyare well. This decision must be clearly recorded in the patient’snotes, so that all those seeing this patient are aware of herdecision. Patients with an asymptomatic rise in CA125 are anideal group to test minimally toxic agents, such as tamoxifen,that might slow tumour progression [6]. If such trials areavailable locally, patients should be given information aboutthem as they might then wish to have regular CA125measurements, so that they can enter a trial where an entrycriterion is a rising CA125 level. The ideal group of patients arethose who have already relapsed and responded to their relapsetherapy. These patients know they will relapse again and arekeen to enter trials that might delay their next relapse. A trial isunderway in which patients start tamoxifen once their CA125levels have reached four times the nadir level (http://clinicaltrials.gov/show/NCT00305838). It is postulated that theslope of the rising CA125 levels before starting an active agentshould be greater than after starting the agent. It is hoped thatfewer patients would be required to show that a drug cansignificantly prolong the doubling time of CA125 than wouldbe required in a randomized trial.Progression-free survival is increasingly used as an endpoint

in clinical trials. Now we know that early treatment of relapseprovides no survival benefit, we need to address how patientsshould be managed if they have progressed just according toCA125. All trials should now give patients the choice of notknowing their CA125 measurements. Where patients are onmaintenance therapy, providing it is well tolerated theyshould be allowed to continue until either a defined timepointor until symptomatic progression. This is particularlyimportant for drugs such as bevacizumab, which has beenshown to delay progression-free survival [7, 8]. The definitionof progression, either by RECIST or CA125, indicatesprogression when precise criteria such as a 20% increase intumour dimension or 50% increase in CA125 level have beenreached. An agent that is delaying that event might well still beworking at the date of designated progression and ifcontinued could further delay symptomatic progression.A good case could therefore be made for not doing regularCT scans and CA125 levels during maintenance therapy, asthey could harm patients by leading to the prematurestopping of an active agent.The OV05/55955 trial shows that patients having 3-monthly

CA125 measurements during follow-up have CA125-definedprogression 4.8 months earlier than those not having routineCA125 measurements. This has major implications if the timefrom last platinum chemotherapy to relapse is used to defineplatinum resistance. Thus, a patient who has CA125-definedprogression at 4 months might well not develop symptoms andthen have RECIST-defined progression until 8 or 9 monthsfollowing their last platinum therapy (Figure 2). Thus, centreswhere routine CA125 measurements are performed are likely tohave patients entering platinum-resistant trials who would beconsidered platinum sensitive in centres where no routineCA125 measurements are performed. The latest consensus

meeting of the Gynecologic Cancer InterGroup tried to resolvethis issue by insisting that it should be stated for each patientwhether the relapse was detected first by RECIST or CA125 andwhether symptoms were present. Trial protocols could thenspecify whether asymptomatic patients with an elevated CA125level are eligible.

conclusion

The lack of benefit from earlier treatment with its deleteriouseffect on quality of life should lead everyone to questionwhether they should alter their follow-up practice. It is a fact oflife that many people will find any excuse to resist change.Some doctors will find aspects of the OV05/55955 trialunsatisfactory; however, none of the published criticisms areserious enough to undermine the main conclusions of the trial[9–11]. Some will say that patients insist on having regularblood tests. Some patients will, but in my practice, followinginformation about OV05/55955 results and adequatecounselling, the great majority now prefer not to have routineCA125 measurements.

disclosures

The author has declared no conflicts of interest.

references

1. Oskay-Oezcelik G, du Bois A, Fasching PA et al. What do patients think about

CA-125 monitoring in the follow-up? Results form a multi-center trial in 1060

patients with ovarian cancer. J Clin Oncol 2009; 27 (Suppl 15): Abstract 5522.

2. Rustin GJ. What surveillance plan should be advised for patients in remission

after completion of first-line therapy for advanced ovarian cancer? Int J Gynecol

Cancer 2010; 20 (Suppl 2): S27–28.

3. Rustin GJS, van der Burg MEL, Griffin CL et al. Early versus delayed treatment of

relapsed ovarian cancer (MRC OV05/EORTC 55955): a randomised trial. Lancet

2010; 376: 1155–1163.

4. Goto T, Takano M, Watanabe A et al. Potential survival benefit of secondary

cytoreductive surgery for recurrent epithelial ovarian, tubal, and peritoneal

cancers. Int J Gynecol Cancer 2011; 21: 263–268.

Figure 2. Example of treating a patient because of their CA125 rising to above

the upper limit of normal (ULN) (point B), when they are 4 months from

completion of their last platinum chemotherapy and considered platinum

resistant, or waiting until they develop symptoms at 8months (point A) and are

considered platinum sensitive. It also shows that more cycles of chemotherapy

are given if chemotherapy is started just because of an elevated CA125.

Annals of Oncology symposium article

Volume 22 | Supplement 8 |December 2011 doi:10.1093/annonc/mdr471 | viii47

by guest on January 3, 2014http://annonc.oxfordjournals.org/

Dow

nloaded from

5. Harter P, du Bois A, Hahmann M et al. Surgery in recurrent ovarian cancer: the

Arbeitsgemeinschaft Gynaekologische Onkologie (AGO) DESKTOP OVAR trial.

Ann Surg Oncol 2006; 13: 1702–1710.

6. Hurteau JA, Brady MF, Darcy KM et al. Randomized phase III trial of tamoxifen versus

thalidomide in women with biochemical-recurrent-only epithelial ovarian, fallopian

tube or primary peritoneal carcinoma after a complete response to first-line platinum/

taxane chemotherapy with an evaluation of serum vascular endothelial growth factor

(VEGF): a Gynecologic Oncology Group study. Gynecol Oncol 2010; 119: 444–450.

7. Burger R, Brady M, Bookman M et al. Phase III trial of bevacizumab (BEV) in the

primary treatment of advanced epithelial ovarian cancer (EOC), primary

peritoneal cancer (PPC), or fallopian tube cancer (FTC): a Gynecologic Oncology

Group study. J Clin Oncol 2010; 28 (18s): (LBA1).

8. Perren T, Swart AM, Pfisterer J et al. ICON 7: a phase III randomised

Gynecologic Cancer InterGroup trial of concurrent bevacizumab and

chemotherapy followed by maintenance bevacizumab, versus

chemotherapy alone in women with newly diagnosed epithelial ovarian

(EOC), primary peritoneal (PPC) or fallopian tube cancer (FTC). Ann Oncol

2010; 21: VIII2.

9. Schwartz PE. Fifth Alon Dembo Memorial Workshop: case 3. Int J Gynecol

Cancer 2010; 21: (Suppl 8), (LBA4) viii 2.

10. Morris RT, Monk BJ. Ovarian cancer: relevant therapy, not timing, is paramount.

Lancet 2010; 376: 1120–1122.

11. Rustin G, van der Burg M, Griffin C et al. Early versus delayed treatment of

relapsed ovarian cancer. Lancet 2011; 377: 380–381.

symposium article Annals of Oncology

viii48 | Rustin Volume 22 | Supplement 8 |December 2011

by guest on January 3, 2014http://annonc.oxfordjournals.org/

Dow

nloaded from