Annals of Oncology Advance Access originally published online on September 26, 2008
Annals of Oncology 2009 20(2):294-297; doi:10.1093/annonc/mdn601
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
gynecologic tumors |
Risk of recurrence during follow-up for optimally treated advanced epithelial ovarian cancer (EOC) with a low-level increase of serum CA-125 levels
1 Department of Medical Oncology
2 Department of Gynecologic Oncology
3 Department of Pathology, Vall d'Hebron University Hospital, Barcelona, Spain
* Correspondence to: Dr A. Prat, Department of Medical Oncology, Vall d'Hebron University Hospital, Passeig de la Vall d'Hebron, 119-129, 08035 Barcelona, Spain. Tel: +3434893000; Fax +3432746059; E-mail: prat.aleix{at}terra.es
 |
abstract |
|---|
 Top abstract introductionpatients and methodsresultsdiscussionreferences |
|---|
Background: Our group evaluated the risk of recurrence for optimally treated advanced epithelial ovarian cancer (adEOC) in patients with a low-level rising serum CA-125 concentration within the normal range (0–35 kU/l). In addition, we tested the new proposed early CA-125 signal of progressive disease (EPD) criterion in the same study population.
Patients and methods: Patients treated from 1998 to 2006 for adEOC were identified at our institution. Inclusion criteria were as follows: CA-125 at time of diagnosis (>35 kU/l); International Federation of Gynecology and Obstetrics stages III–IV treated with optimal primary treatment; and complete response (CR) to primary treatment with normalization of CA-125.
Results: Median progression-free survival and overall survival for the recurrence group (n = 60) were 17.7 and 38.2 months, respectively. The median follow-up time from CR to last contact was 40.2 months for patients in the nonrecurrence group (n = 36). An absolute increase in serum CA-125 levels of 
5 kU/l compared with baseline CA-125 nadir values was significantly predictive of recurrence (odds ratio for recurrence = 402.98, P < 0.0001). The progression date was predated by the EPD criterion in 77% of patients with known progressive disease (median, 58 days early) with a sensitivity of 90%, a positive predictive value of 96.4%, and a false-positive rate of 5.6%.
Conclusions: Among patients with optimally treated adEOC in complete remission, a low-level increase in serum CA-125 concentration within the normal range is a strong independent predictive factor for disease recurrence. In this patient population, future prospective randomized trials should consider the evaluation of the EPD criterion.
Key words: CA-125, ovarian cancer, recurrence, residual disease, surveillance
|
introduction |
|---|
|
Topabstractintroductionpatients and methodsresultsdiscussionreferences |
|---|
Despite the performance of an optimal primary treatment for patients with advanced epithelial ovarian cancer (adEOC), recurrences occur in the majority of cases and low survival rates are expected (30%–50% at 5 years) [1, 2]. Consequently, there is considerable interest in trying to identify early indicators of relapse for this patient population for optimizing therapeutic management [3, 4].
The serum CA-125 concentration has been established as a tool of great importance during surveillance for EOC recurrence. The elevation of this serum tumor marker usually antedates the development of a clinical relapse with a median duration of at least 4.5 months (range 0.5–29.5 months) [5]. For many years, clinical trials have defined progressive disease after a complete response (CR) complete clinical and radiographic response with normalization of serum CA-125 (
35 kU/l) to primary therapy on the basis of the response evaluation criteria in solid tumors (RECIST) criteria and/or the Gynecologic Cancer Intergroup (GCIG) CA-125 criteria [6]: an elevation of CA-125 to two-fold (2x) the upper limit of normal (ULN, 35 kU/l) documented on two occasions. However, in this group of patients, it has recently been reported that the CA-125 nadir levels within the normal range (35 kU/l) are an independent prognostic factor for progression-free survival (PFS) [7–11]. Indeed, CA-125 values between 11 and 35 kU/l have been associated with a decrease in PFS compared with the group of patients who achieve nadir values of 10 kU/l [7–11]. Furthermore, data from one report suggest that an increase in CA-125 serum concentration within the normal range is a strong predictive factor for disease recurrence [12].
On the basis of these findings, a new CA-125 criterion for progressive disease has been proposed by Liu et al. [13] that could better reflect the timing of disease progression in patients with CR after primary treatment. The new early signal of progressive disease (EPD) criterion has been defined as a value of 20 kU/l for patients with CA-125 nadir 10 kU/l and a value 2x nadir for patients with nadir values >10 kU/l. Tested in SWOG-9701/GOG-178 trial [14], which evaluated maintenance paclitaxel in patients with CR after primary treatment for adEOC, the EPD criterion demonstrated a low false-positive rate, a higher sensitivity, and an early prediction of EOC progression when compared with the RECIST/GCIG CA-125 criteria.
Thus, the aim of our retrospective study is two-fold. First, to determine the risk of relapse on the basis of low absolute changes in CA-125 serum levels within the normal range, and, secondly, to test the EPD criterion in a homogeneous study population of patients with optimally treated adEOC.
|
patients and methods |
|---|
|
Topabstractintroductionpatients and methodsresultsdiscussionreferences |
|---|
patient populations
The retrospective identification of our patient population has been published elsewhere [11]. Briefly, from 1 January 1998 to 31 December 2006, 96 patients with adEOC [International Federation of Gynecology and Obstetrics (FIGO) stage of diseases III–IV] were identified from the tumor registry database at Vall d'Hebron University Hospital. All patients underwent optimal cytoreductive surgery (residual tumor <1 cm in maximal diameter) with preoperative and/or postoperative platinum- and taxane-based combination chemotherapy. All patients achieved a complete clinical, radiological, and biochemical response (
35 kU/l) to primary treatment and were followed up by observation only. The general clinical practice for surveillance of adEOC patients at our institution was to perform a complete physical examination, determination of serum CA-125 concentrations, and computed tomography (CT) scanning of the thorax, abdomen, and pelvis at 2- to 4-month intervals for the initial 36 months after achieving a CR. For an additional 24 months, clinical examinations, determination of serum CA-125 concentrations, and CT scans were obtained at 6-month intervals. Thereafter, follow-up was undertaken every year for at least an additional 36 months. If recurrence was clinically suspected, the above studies or others such as positron emission tomography (PET) were conducted at any time during surveillance. This retrospective study was approved by the ethics committee of our institution.
statistical methodology
The primary study goal was to determine whether a low increase (5 kU/l) in CA-125 serum concentrations within the normal range is an independent predictor of radiological EOC progression/recurrence. Disease progression/recurrence was defined on imaging-based criteria: appearance of new lesions by CT scan or by PET scan imaging. PFS was defined as the interval between the date of primary surgery or the date of initiation of neo-adjuvant chemotherapy and the date of progression/recurrence or death, whichever came first. Overall survival (OS) was calculated from the date of primary surgery or the date of initiation of neo-adjuvant chemotherapy until the last date of follow-up or date of death.
To predict the risk of recurrence, a multivariate regression model adjusted for stage and CA-125 nadir group was used to compare absolute changes in CA-125 levels among patients with and without EOC recurrence. Patients with the first detected CA-125 elevation above the ULN (n = 22) were excluded from the regression analysis. Absolute changes were calculated by subtracting the CA-125 baseline nadir after primary treatment from the CA-125 level at any time during surveillance. The median lead time before disease progression was calculated by subtracting the date of progression/recurrence from the first date of occurrence of a low-level increase in CA-125 from baseline nadir or the occurrence of the EPD criterion, whichever applicable. Bivariate correlation analysis was used to assess the correlation between the time to occurrence of the EPD criterion and OS. The patient characteristics of the recurrence and nonrecurrence groups were compared by student's t-test for continuous variables and by Fisher's exact test or Pearson chi-square test for discrete variables. Results were considered significant for P < 0.05.
|
results |
|---|
|
Topabstractintroductionpatients and methodsresultsdiscussionreferences |
|---|
After a median follow-up of 38.6 months (range 2–96 months), 60 patients (62.5%) presented a recurrence and 36 patients (37.5%) remained disease free. The median follow-up of the nonrecurrence group was 40.2 months (range 2–85 months). For the recurrence group, median PFS and OS were 17.7 months (range 7–53 months) and 38.2 months (range 10–96 months), respectively. Baseline characteristics were well balanced in both groups (see Table 1) with the exception of median serum CA-125 nadir levels being higher in the recurrence group, which is consistent with previous findings [11].
|
A low-level absolute increase (
5 kU/l) in CA-125 levels within the normal range was observed in 40 patients (37 patients in the recurrence group versus three patients in the nonrecurrence group), and the median lead time from the occurrence of a low-level increase in CA-125 within the normal range to disease progression was 3.0 months (range 0–9 months). Among the group of patients with known progressive disease, the median serum concentration after a low-level increase in CA-125 was 25.5 kU/l (range 11–35 kU/l). Conversely, in the nonrecurrence group, the median CA-125 level at last contact was 10.6 kU/l (range 5–32 kU/l), which did not differ substantially from the median baseline CA-125 nadir level (9.67 kU/l). In multivariate logistic regression analysis, only the occurrence of a low-level absolute increase from baseline CA-125 nadir was a highly statistically significant predictor of EOC relapse (odds ratio for recurrence = 402.98; 95% confidence interval 39–4112; P < 0.0001) (see Table 2).
|
In a subsequent analysis, we tested the recently defined EPD criterion in our study population (see Table 3). Fifty-six patients met the proposed CA-125 EPD criterion at a median follow-up time of 15 months (range 5–53 months): 54 patients in the recurrence group (sensitivity, 54 of 60, 90%) and two patients in the nonrecurrence group (false-positive rate, 2 of 36, 5.6%). In addition, six patients relapsed without meeting the EPD criterion (false-negative rate, 6 of 60, 10%). The EPD criterion gave an early signal in 77% of patients with known progressive disease, predating the radiological progression/recurrence in 42% by 60 days, 20% by 61–180 days, and 15% by >180 days, with a median lead time from EPD occurrence to disease progression of 58 days. The median CA-125 level at occurrence of EPD was 47.5 kU/l (range 20–325 kU/l), and 64% of patients who met the EPD criterion presented a CA-125 value <70 kU/l (GCIG CA-125 criteria). Finally, bivariate correlation analysis revealed a strong positive correlation between the time to occurrence of the EPD criterion and OS [Spearman's correlation coefficient (r2) = 0.62, P < 0.0001].
|
|
discussion |
|---|
|
Topabstractintroductionpatients and methodsresultsdiscussionreferences |
|---|
Several reports have recently suggested that it may be possible to define the risk of relapse and death by dividing patients who have achieved a complete biochemical (
35 kU/l) and radiological response after primary treatment into arbitrary groups based on the CA-125 nadir (10 versus 11–20 versus 21–35 kU/l) [7–11]. Indeed, CA-125 nadir levels of 10 kU/l have been associated with longer relapse-free survival and OS compared with patients with nadir values between 11 and 35 kU/l [11]. Given the differences in prognosis between the CA-125 nadir groups, the predictive role of an increase in CA-125 nadir levels within the normal range (0–35 kU/l) was evaluated previously in a heterogeneous group of 39 patients (stages I–IV) who achieved a CR to primary treatment [12]. Univariate logistic regression analyses showed that an absolute increase of 5 kU/l, 10 kU/l, or a doubling of the CA-125 nadir within the normal range were strong predictive factors for disease recurrence. However, there were certain limitations to this study secondary to the small sample size and the heterogeneity of the study groups with respect to the histological subtype and FIGO stage. To our knowledge and despite analyzing retrospective data, our report is the first to support the predictive role for relapse of a low-level increase of serum CA-125 in a homogeneous group of patients with optimally treated adEOC even when other variables are taken into consideration.
What can be suggested from our results? First, disease recurrence continues to be a major issue even in optimally treated patients with adEOC that have achieved a complete clinical and biochemical response after primary treatment. In our population group, 60 (62.5%) patients manifested EOC recurrence and 50% of them succumbed to their disease at 3.2 years from diagnosis. Thus, it would be tremendously interesting to find early signs of progression/recurrence in order to optimize new treatment strategies. One of these early signs might be a low-level increase in CA-125 serum values within the normal range during surveillance for EOC recurrence. Indeed, an absolute increase of 5 kU/l as discriminating criteria identified 97.4% of patients in our study population with recurrent disease and 92% of patients who remained disease free. Consequently, the false-positive prediction of relapse was 8%, with a sensitivity of 97.4%, a specificity of 92%, and a positive predictive value (PPV) of 92.5%. These results compared satisfactory to the previous results reported by Santillan et al. [12].
Secondly, the optimal surveillance strategy for recurrence has not been properly defined in this group of patients. Our general clinical practice has been to perform serial CT scans and CA-125 measurements as described in ‘patients and methods’. However, the high negative predictive value of our discriminating criteria (97.1%) suggests that asymptomatic patients without a low-level increase in CA-125 values within the normal range have a low probability of disease recurrence and questions the need for further imaging studies that would expose patients to unnecessary radiation.
Thirdly, the long median interval time of 3.0 months (range 0–9 months) from the time of an absolute increase of >5 kU/l within the normal range to the time of documented recurrent disease by imaging studies can be explained by the institutional practice of not performing imaging studies at an earlier interval until CA-125 serum values above the ULN are identified on two successive occasions at least 4 weeks apart.
Fourthly, for the group of patients with CR after primary treatment, the GCIG CA-125 criteria predicts tumor relapse with a sensitivity of 84% and a false-positive rate of <2% [15]. However, given that in the last 2 years different prognostic groupings have been described within the normal CA-125 range [7–11], and that increases of CA-125 levels within this range are predictive for recurrence, a new CA-125 criterion for progression/recurrence that takes into consideration these findings might be more appropriate in this group of patients. Liu et al. [13] has recently proposed an alternative CA-125 criterion for disease relapse, known as the EPD. The new EPD criterion is based on the GCIG CA-125 disease progression criterion, but with a lower ULN from 35 to 10 kU/l. Tested on the SWOG-9701/GOG-178 trial [14] and compared with the RECIST/GCIG CA-125 criterion, the progression date was predated by EPD by a median of 1.9 months (56 days), and the PPV of the proposed criterion was 93%. Using this more sensitive indicator, the authors found that >50% of patients would have had an earlier date of progression. When we applied the EPD and made a comparison with the imaging-based criteria in our homogeneous study population, the radiographic disease progression was predated by EPD in 77% patients by a median of 58 days with high sensitivity and low false positivity concordant with previous findings [13]. The two patients from the nonrecurrence group who satisfied the EPD criterion (false positives) had a median follow-up from CA-125 nadir value of 23 and 32 months, respectively. Both patients satisfied the EPD criterion at their last follow-up contact. Interestingly, the first patient had documented disease recurrence 5 months after completion of the study. The second patient has continued surveillance for 6 months without evidence of recurrence (38 months total from CA-125 nadir).
Fifthly, although the utility of serial measurements of serum CA-125 in the surveillance for recurrence, the impact on survival of detecting and treating earlier asymptomatic disease recurrence is not known at the present time [16]. Consequently, one prospective randomized trial is being conducted by the Medical Research Council/EORTC (EORTC protocol 55955/OV05) in Europe that will evaluate whether or not early intervention favorably affects outcome. Patients with EOC with complete remission after primary treatment will be randomized, based on GCIG CA-125 criteria alone to determine whether early chemotherapy treatment can improve symptom-free survival, OS, and quality of life.
Finally, our results suggest that a low-level increase in CA-125 serum levels within the normal range is a strong independent predictive factor for disease progression/recurrence in optimally treated patients with adEOC who have achieved a CR. In this patient population, the proposed EPD criterion may better reflect the timing of disease progression and should be investigated further in prospective randomized clinical trials.
Received for publication October 25, 2007. Revision received February 5, 2008. Accepted for publication August 4, 2008.
|
references |
|---|
|
Topabstractintroductionpatients and methodsresultsdiscussionreferences |
|---|
1. Paclitaxel plus carboplatin versus standard chemotherapy with either single-agent carboplatin or cyclophosphamide, doxorubicin, and cisplatin in women with ovarian cancer: the ICON3 randomised trial. Lancet (2002) 360:505–515.[CrossRef][Web of Science][Medline]
2. Ozols RF, Bundy BN, Greer BE, et al. Phase III trial of carboplatin and paclitaxel compared with cisplatin and paclitaxel in patients with optimally resected stage III ovarian cancer: a Gynecologic Oncology Group Study. J Clin Oncol (2003) 21:3194–3200.
3. Riedinger JM, Bonnetain F, Basuyau JP, et al. Change in CA 125 levels after the first cycle of induction chemotherapy is an independent predictor of epithelial ovarian tumour outcome. Ann Oncol (2007) 18:881–885.
4. Riedinger JM, Wafflart J, Ricolleau G, et al. CA 125 half-life and CA 125 nadir during induction chemotherapy are independent predictors of epithelial ovarian cancer outcome: results of a French multicentric study. Ann Oncol (2006) 17:1234–1238.
5. van der Burg MEL, Lammes FB, Verweij J. The role of CA 125 in the early diagnosis of progressive disease in ovarian cancer. Ann Oncol (1990) 1:301–302.
6. Vergote I, Rustin GJS, Eisenhauer EA, et al. Re: new guidelines to evaluate the response to treatment in solid tumors [ovarian cancer]. J Natl Cancer Inst (2000) 92:1534–1535.
7. Crawford SM, Paul J, Reed NS, et al. The prognostic significance of the CA125 nadir in patients that achieve a CA125 response. Proc Am Soc Clin Oncol (2004) 23:448. Abstr 5001.
8. Crawford SM, Peace J. Does the nadir CA125 concentration predict a long-term outcome after chemotherapy for carcinoma of the ovary? Ann Oncol (2005) 16:47–50.
9. Markman M, Liu PY, Rothenberg ML, et al. Pretreatment CA-125 and risk of relapse in advanced ovarian cancer. J Clin Oncol (2006) 24:1454–1458.
10. Juretzka MM, Barakat RR, Chi DS, et al. CA125 level as a predictor of progression-free survival and overall survival in ovarian cancer patients with surgically defined disease status prior to the initiation of intraperitoneal consolidation therapy. Gynecol Oncol (2007) 104:176–180.[CrossRef][Web of Science][Medline]
11. Prat A, Parera M, Peralta S, et al. Nadir CA-125 concentration in the normal range as an independent prognostic factor for optimally treated advanced epithelial ovarian cancer. Ann Oncol (2007).
12. Santillan A, Garg R, Zahurak ML, et al. Risk of epithelial ovarian cancer recurrence in patients with rising serum CA-125 levels within the normal range. J Clin Oncol (2005) 23:9338–9343.
13. Liu P-Y, Alberts DS, Monk BJ, et al. An early signal of CA-125 progression for ovarian cancer patients receiving maintenance treatment after complete clinical response to primary therapy. J Clin Oncol (2007) 25:3615–3620.
14. Markman M, Liu PY, Wilczynski S, et al. Phase III randomized trial of 12 versus 3 months of maintenance paclitaxel in patients with advanced ovarian cancer after complete response to platinum and paclitaxel-based chemotherapy: a Southwest Oncology Group and Gynecologic Oncology Group Trial. J Clin Oncol (2003) 21:2460–2465.
15. Rustin GJS, Nelstrop AE, Tuxen MK, Lambert HE. Defining progression of ovarian carcinoma during follow-up according to CA 125: a North Thames Ovary Group study. Ann Oncol (1996) 7:361–364.
16. Goonewardene TI, Hall MR, Rustin GJS. Management of asymptomatic patients on follow-up for ovarian cancer with rising CA-125 concentrations. Lancet Oncol (2007) 8:813–821.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||