Skip to main content
Free access
Gynecologic Cancer
August 15, 2004

Early Detection and Prognosis of Ovarian Cancer Using Serum YKL-40

Publication: Journal of Clinical Oncology

Abstract

Purpose

YKL-40 is a secreted glycoprotein (chitinase family). We compared YKL-40 with two ovarian cancer serum markers, CA125 and CA15-3, for the detection of early-stage ovarian cancer.

Materials and Methods

Serum YKL-40 levels were assayed by enzyme-linked immunosorbent assay for 46 healthy subjects, 61 high-risk individuals, 33 patients with benign gynecologic processes, and 50 preoperative patients subsequently diagnosed with predominantly early-stage ovarian cancer. Serum CA125 and CA15-3 values were obtained.

Results

Median YKL-40 level was 28 ng/mL (range, 15 to 166 ng/mL) for healthy subjects, 36 ng/mL (range, 9 to 69 ng/mL) for high-risk individuals without prior cancer, 44.5 ng/mL (range, 5 to 133 ng/mL) for high-risk patients with prior breast cancer, and 38 ng/mL (range, 5 to 67 ng/mL) for individuals with benign gynecologic processes (P = NS). Median preoperative YKL-40 level for ovarian cancer patients was 94 ng/mL (range, 17 to 517 ng/mL; P < .0001 compared with normal and high-risk). YKL-40 was elevated (≥ 62 ng/mL) in 36 (72%) of 50 patients compared with 23 (46%) of 50 and 13 (26%) of 50 patients for CA125 and CA15-3 (P < .008). Twenty (65%) of 31 early-stage patients had elevated serum YKL-40 levels compared with 11 (35%) of 31 and four (13%) of 31 patients for CA125 and CA15-3 (P = .039). YKL-40 levels increased with stage (P < .005), regardless of grade, histology, or patient age. Patients with early-stage tumors with YKL-40 values more than 80 ng/mL had a worse prognosis (71% recurrence v no recurrence [P = .034]).

Conclusion

YKL-40 may represent a novel marker for the detection of early-stage ovarian cancer. YKL-40 levels in early-stage patients may also predict disease recurrence and survival. The utility of YKL-40 in detection of early-stage ovarian cancer deserves further investigation.

Introduction

Ovarian cancer is the leading cause of death from gynecologic malignancies in the United States. Over 23,000 cases are diagnosed yearly, and there are an estimated 14,000 deaths per year due to ovarian cancer.1 More than 70% of patients have stage III or IV disease at the time of diagnosis.2 Despite the introduction of new chemotherapeutic agents for the treatment of ovarian cancer, the fact that most patients are diagnosed with advanced-stage disease translates into a poor 5-year survival of 20% to 30%.2 On the other hand, 90% of women diagnosed with disease confined to the ovary survive more than 5 years. Thus, early detection of ovarian cancer is essential for improved survival. The development of serum-based diagnostic tests for the detection of early-stage cancers in asymptomatic patients has been an important endeavor. Unlike breast or prostate cancer, there is no simple diagnostic test to detect early-stage ovarian tumors. Despite considerable efforts, no cost-effective screening tests have been developed. There is a need for new diagnostic markers of ovarian cancer that might be useful for screening purposes.
Currently, a number of ovarian cancer markers have been defined. These include: CA125, CA15-3, carcinoembryonic antigen, the kallikrein family (including hK13, hK10, and hK6), prostasin, lysophosphatidic acid (LPA), and others.3 CA125 was the first tumor marker available for the management of ovarian cancer and remains the best. Elevation of CA125 in the serum can be detected in benign conditions, including pregnancy, endometriosis, ovarian cysts, and cirrhosis, and in malignant conditions, such as ovarian, fallopian tube, primary peritoneal, cervical, endometrial, breast, colon, and lung cancers. Approximately 80% of patients with epithelial ovarian cancer have elevations of CA125.4 CA125 is useful for disease monitoring, assessing response to therapy, and the detection of relapse.5-7 The major problems with CA125 serum marker include poor sensitivity and specificity for ovarian cancer, especially for the diagnosis of early-stage disease. CA125 is elevated in only 40% to 50% of patients with stage I/II tumors.8 CA15-3 is elevated in several tumors including breast cancer, and it is elevated in approximately 70% of epithelial ovarian cancer patients,9,10 predominantly in those with advanced disease. The human kallikrein gene family is a subfamily of serine proteases and can be involved in the progression and metastasis of human cancers.11-13 Recent studies have detected the secreted kallikreins 6 and 10 in patients with ovarian cancer.14,15 Prostasin has recently been evaluated as another serum marker for nonmucinous ovarian carcinomas,16 as has LPA.17 In a small study, plasma LPA concentrations were elevated in 90% of women with stage I disease and 100% of women with advanced and recurrent disease compared with healthy controls. However, the current method of measuring LPA, which involves lipid extraction followed by gas chromatography, may limit its utility.
YKL-40 is a glycoprotein in the chitinase protein family. The gene for YKL-40 is located on chromosome 1q3218 and is a mammalian member of the 18-glycosyl-hydrolase family,19 a gene family that includes bacterial and fungal chitinases. YKL-40 has significant sequence similarity to the chitin-degrading enzyme, chitinase20; it has been shown to bind chitin, but retains no chitinase activity and has not been determined to have any other enzymatic activity or function. The full spectrum of mammalian polysaccharide structures that bind to YKL-40 and the function of this protein are unknown.
It has been reported that YKL-40 is expressed in pathologic conditions of extracellular matrix degradation and angiogenesis. It has been evaluated as a serum marker for conditions such as rheumatoid arthritis,21 severe osteoarthritis,22 hepatic fibrosis, primary colorectal cancer,23 glioblastoma,24 metastatic breast cancer,25 and recurrent ovarian cancer.26 These data suggest YKL-40 may be involved in extracellular matrix degradation and/or angiogenesis.27,28
We initiated this study of ovarian cancer patients to determine if YKL-40 protein is elevated in the serum of early-stage patients. The purpose of this study was to determine if YKL-40 represents a possible serum marker for early-stage ovarian cancer that might be useful for early detection and prognostic purposes.

Materials and Methods

Study Population

Patients undergoing gynecologic oncologic surgery at Memorial Sloan-Kettering Cancer Center (MSKCC; New York, NY) had their tumor specimens and serum samples banked under an institutional review board- (IRB-) approved tissue-acquisition protocol after signing informed consent. MSKCC pathologists reviewed all tumor specimens. Patients in the study were operated on between 1992 and 2002. Patients were selected based on a diagnosis of ovarian, fallopian tube, or primary peritoneal cancer and were preferentially selected if they were diagnosed with an early-stage cancer. In addition, the first 31 patients identified with stage I and II cancer were included in the study. Patients also had to have no prior history of arthritis or other malignancy in order to be eligible. Patient identifiers were removed from the samples, and the MSKCC IRB approved the study.
Twenty-seven anonymous serum samples from normal individuals were obtained from the blood donor room at MSKCC, and 19 previously defined normal samples4 were selected. These individuals were not screened for pre-existing conditions such as arthritis or malignancy. Serum samples were also obtained, with IRB approval, from 61 individuals enrolled in the high-risk ovarian cancer screening program.29 High-risk individuals were also followed with CA125 serum marker and pelvic sonogram. Serum samples tested for YKL-40 were from all individuals who had been followed for at least 1 year in the high-risk screening program without evidence of ovarian cancer. Serum samples from 33 individuals without known cancer but with benign gynecologic disease such as uterine fibroids, ovarian cysts, endometriosis, and endometrial hyperplasia were identified in the MSKCC gynecologic tissue bank.

Blood Collection and Serum Separation

All blood samples were collected from patients with ovarian, fallopian tube, or primary peritoneal cancer 1 to 2 weeks before primary surgical resection of their tumors. The blood samples from patients, normal individuals, and high-risk individuals were allowed to clot at room temperature for at least 30 minutes, and were then centrifuged at 4°C for 5 minutes at 1,500 rpm. The serum aliquots were stored at −20°C until tested.

YKL-40 Enzyme-Linked Immunosorbent Assay (ELISA)

YKL-40 levels were determined, in duplicate, for all serum samples, using the commercially available YKL-40 ELISA Kit from Metra Biosystems (Mountain View, CA), according to the manufacturer's protocol. Protein concentrations were determined as absorbances using the Bio-Rad Benchmark Microplate Reader (Bio-Rad Laboratories, Hercules, CA).

CA125 and CA15-3 Serum Analysis

CA125 and CA15-3 serum testing was performed in the clinical chemistry laboratory of MSKCC, on an Immuno 1 analyzer from Bayer Diagnostics (Tarrytown, NY). For data analysis, the upper limit of normal for CA1254 and CA15-39 were defined as 35 U/mL.

Statistical Analysis

To analyze the data, we divided patients into different groups according to clinical and pathologic parameters. Comparisons between groups were done using Student's t test, after applying the logarithmic transformation to reduce the skewness in the distribution of the laboratory assay results. Correlations between numeric variables were assessed by Spearman rank-correlation coefficient. YKL-40, CA125, and CA15-3 serum concentrations were also classified as normal or elevated. The statistical significance of the relative accuracy of YKL-40 versus that of either CA125 or CA15-3 in detecting cancer among subgroups of patients was based on cases where two tests gave discrepant results using McNear's test. The relationship of these dichotomous variables to other clinicopathologic correlates was established with the χ2 test or the Fisher's exact test, as appropriate.
Receiver operating characteristics curves (ROC) were constructed for YKL-40 and CA125 serum concentrations as diagnostics for cancer by plotting sensitivity versus 1-specificity, and the area under each ROC curve was calculated. Statistical analysis comparing the ROC curves of YKL-40 and CA125 was performed using the method of DeLong et al.30 A Kaplan-Meier progression-free survival curve was constructed to demonstrate, in stage I and II patients, the progression-free survival difference between patients with YKL-40 elevations less than and greater than 80 ng/mL. The log-rank test was used to examine the significance of the relationship between log-marker values and progression-free survival.

Results

Normal Subject YKL-40, CA125, and CA15-3 Values

Serum was collected from 46 normal subjects. As depicted in Table 1, the range of YKL-40 levels in the normal patients was 15 to 166 ng/mL. The mean and median YKL-40 values were 33.5 ng/mL and 28 ng/mL, respectively. The mean value is virtually identical to the mean serum YKL-40 value of 33 ng/mL obtained for 102 healthy women in another recent publication.26
The upper limit of normal for YKL-40 in this group of normal individuals was defined as 61 ng/mL, based on the mean value plus two standard deviations (95% CI). Thus, an abnormal YKL-40 serum level was determined to be ≥ 62 ng/mL. This value is consistent with the reagent vendor (Metra Biosystems). Four of 46 individuals had YKL-40 values ≥ 62 ng/mL; these values were 166, 140, 72, and 62 ng/mL. For the sake of confidentiality, normal individuals were not questioned about a personal history of arthritis or cancer.
CA125 mean and median values were 13.4 U/mL and 11.5 U/mL (range, 4 to 31 U/mL), respectively. CA15-3 mean and median values were 17 U/mL and 15.5 U/mL (range, 7 to 34 U/mL).

High-Risk Screening Patient YKL-40, CA125, and CA15-3 Values

Serum was collected from patients followed for more than 1 year in the high-risk ovarian cancer-screening program at MSKCC. Nineteen patients had no personal history of malignancy. Forty-two patients had a personal history of breast cancer for which they were without evidence of disease. All patients had a strong family history of breast or ovarian cancer (Table 1).
The median YKL-40 values were 44.5 ng/mL (range, 5 to 133 ng/mL) and 36 ng/mL (range, 9 to 69 ng/mL) for high-risk individuals with and without a personal history of breast cancer, respectively. There was no statistically significant difference between these groups or between either the high-risk group and the normal individuals tested. For the patients with no prior history of cancer, one patient had a YKL-40 value of 69 ng/mL. For the patients with a prior history of breast cancer, 11 patients had YKL-40 values greater than 61 ng/mL. Of note, three of these patients, with YKL-40 values of 87 ng/mL, 87 ng/mL, and 62 ng/mL, were found 18, 8, and 18 months later, respectively, to have new pulmonary nodules, recurrent breast cancer, and an abdominal mass on a computed tomography scan, respectively. CA125 values for all screening patients were less than 35 U/mL. Interestingly, there was a difference between mean CA125 values in screening patients with prior cancer and those without prior cancer (P < .001), and between screening patients with prior cancer and normal patients (P < .015).

Patients With Benign Gynecologic Processes

Individuals with benign gynecologic processes based on transvaginal sonogram and pathology reports were identified from the high-risk ovarian screening program. Thirty-three individuals were identified. Diagnoses included uterine fibroids (16 patients), simple ovarian cysts (10 patients), complex ovarian cysts (six patients), corpus luteum cysts (three patients), endometrial polyps (two patients), atypical endometrial hyperplasia (two patients), and endometriosis (one patient).
For the patients with benign gynecologic disorders, the median YKL-40 value was 38 ng/mL (range, 5 to 67 ng/mL), and median CA125 value was 12.5 U/mL (range, 5 to 274 U/mL; Table 1). There was no statistically significant difference between the YKL-40 values of this group and the high-risk groups or the normal individuals tested.
All patients in the benign gynecologic process group had YKL-40 values less than 62 ng/mL except for the two patients with atypical endometrial hyperplasia (YKL-40 values of 62 and 67 ng/mL). All patients in this group had CA125 values less than 35 U/mL except for one patient with endometriosis (CA125 value of 274 U/mL). This patient remains disease-free at 18 months' follow-up.

Ovarian Cancer Patient Study Population

Preoperative serum samples from 50 epithelial ovarian cancer patients were evaluated in this study. Patient demographics are outlined in Table 2. The median age of patients was 59 years (range, 31 to 81 years). Forty-six (92%) of the 50 patients had a diagnosis of primary ovarian cancer. Four patients had primary fallopian tube or peritoneal cancer. Thirty-one (62%) of 50 patients in the study had stage I or II cancers, while the rest had advanced-stage or recurrent disease. Thirty-seven (74%) of the tumors were histologic grade 3. Twenty-two patients (44%) had tumors with serous histology, the most common histologic tumor type. Clinical follow-up was available on 47 (94%) of the 50 patients. Median long-term follow-up of 99 months (range, 33 to 125 months) was available for stage I and II tumors. Thirty-seven (74%) of the patients in the study were alive, and 30 (60%) remained in remission.

Serum YKL-40 Levels in Ovarian Cancer Patients

The mean and median YKL-40 levels in all epithelial ovarian cancer patients were 121.8 ng/mL and 94 ng/mL (range, 17 to 517 ng/mL), respectively (Table 1). As illustrated in Figure 1, preoperative YKL-40 levels were significantly higher in epithelial ovarian cancer patients (P < .0001) relative to both normal controls and individuals in the high-risk epithelial ovarian cancer-screening program. Patients with stage I tumors had preoperative serum YKL-40 levels approximately 2.59 times higher than normal controls (Fig 2A).
Serum values of YKL-40, CA125, and CA15-3 were determined for epithelial ovarian cancer patients in the study (Table 3). For all patients examined, 36 (72%) of 50 epithelial ovarian cancer patients had elevated YKL-40 serum levels compared with 23 (46%) of 50 patients and 13 of 50 patients (26%) for CA125 and CA15-3, respectively. Serum YKL-40 testing was significantly better (P < .008) than CA125 and CA15-3 (P < .0001) at detecting epithelial ovarian cancer when considering patients for whom YKL-40 and CA125 gave discordant results. Nine patients (18%) had elevations of all three markers; 12 patients (24%) had elevations of YKL-40 and CA125; 15 patients (30%) had elevations of YKL-40 alone; one patient (2%) had an elevation in CA125 and CA15-3; two patients (4%) had an elevation in CA125 alone; two patients (4%) had an elevation in CA15-3 alone; and nine patients (18%) did not have elevations in any of the markers.

Serum YKL-40 Levels in Patients With Early-Stage Ovarian Cancer

The mean and median YKL-40 levels in stage I and II ovarian cancer patients were 109 ng/mL and 75 ng/mL, respectively (range, 17 to 517 ng/mL). Preoperative serum levels of YKL-40, CA125, and CA15-3 were elevated in 20 (65%) of 31, 11 (35%) of 31, and four (13%) of 31 stage I and II ovarian cancer patients, respectively. YKL-40 was significantly more likely to detect early-stage ovarian cancer than were CA125 and CA15-3 (P = .039).

Serum YKL-40 Levels in Patients With Advanced and Recurrent Ovarian Cancer

The mean and median YKL-40 levels in 11 advanced-stage ovarian cancer patients were 181 ng/mL and 148 ng/mL (range, 52 to 445 ng/mL), respectively. The mean and median YKL-40 levels in eight recurrent patients were 88 ng/mL and 79 ng/mL (range, 30 to 202 ng/mL), respectively. Preoperative serum levels of YKL-40 were elevated in patients with advanced-stage (10 [91%] of 11 patients) and recurrent (six [75%] of eight patients) ovarian cancer (Table 3). Preoperative serum levels of CA125 were elevated in patients with advanced and recurrent ovarian cancer: seven (64%) of 11 and five (63%) of eight patients for stage III/IV and recurrent tumors, respectively. Preoperative serum levels of CA15-3 were elevated in patients with advanced and recurrent ovarian cancer: six (55%) of 11 and three (38%) of eight patients for stage III/IV and recurrent tumors, respectively.

Frequency of Serum YKL-40 Values in Ovarian Cancer Patients As a Function of Tumor Stage, Grade Histology, and Age

Table 3 enumerates the frequency of elevation of preoperative serum levels of YKL-40, CA125, and CA15-3 in patients who were subsequently diagnosed as having primary ovarian, fallopian tube, or peritoneal cancer on surgical pathologic review. Table 3 further delineates the frequency of elevation of these three serum makers, taking into consideration primary disease site, stage, grade, and histology. The number of patients with elevated serum YKL-40 values was higher than that for CA125 and CA15-3 in all groups, regardless of these variables.
As depicted in Figure 2A, preoperative serum levels of YKL-40 increased with increasing stage of the tumor (stage I v II v III/IV by simple Spearman correlation: r = 0.425; P < .005). Patients with stage II tumors had preoperative YKL-40 values that were 1.39 times higher than those of patients with stage I tumors. Patients with stage III/IV tumors had YKL-40 values that were 1.58 times higher than those of stage II patients. There was a linear trend in YKL-40 elevation when stage I, II, and III/IV patients were compared (P < .005).
Of note, patients with recurrent tumors had elevated, but lower, overall values of serum YKL-40 than patients with newly diagnosed stage II, III, IV tumors. Serum YKL-40 values were approximately 50% lower in patients with pure clear-cell tumors.
Serum YKL-40 values were elevated across all grades of ovarian tumors (Fig 2B). Thirty-seven (74%) of the 50 patients in the study with ovarian tumors had grade 3 tumors; 24 (65%), 17 (46%), and 11 (30%) of these 37 patients had elevated serum YKL-40, CA125, and CA15-3 values, respectively.
Serum YKL-40 values were elevated in all histologic subtypes (Fig 2C). YKL-40 was elevated in 16 (73%) of 22 of patients with serous tumors and in six (75%) of eight patients with endometroid tumors. CA125 and CA15-3 were elevated in 12 (55%) of 22 and 10 (45%) of 22 patients with serous tumors, respectively. Three (38%) of eight patients with endometrioid tumors had elevations of serum CA125, and none of these patients had elevations in CA15-3. Patients with clear-cell tumors were less likely to have elevations of YKL-40 (one of five patients, 20%), CA125 (one of five patients, 20%), and CA15-3 (zero of five patients). YKL-40 was elevated in four of five patients with mucinous tumors, while serum CA125 and CA15-3 were normal in all of these patients. There was no significant correlation between YKL-40 and age overall (r = 0.18; P = .20; Fig 2D).

ROC Curve of YKL-40 in Ovarian Cancer

Figure 3 depicts the ROC curves for YKL-40 and for CA125. Curves were generated for ovarian cancer patients versus normal controls, for ovarian cancer patients versus screening individuals, and for ovarian cancer versus benign gynecologic processes. The approximate area under the ROC curve assessing serum YKL-40 as a diagnostic tool for the detection of ovarian cancer against normal controls was 0.889 compared with 0.782 for CA125 (P = .045; Fig 3). At the value of 61 ng/mL, YKL-40 has a sensitivity and specificity of 72% and 90%, respectively, for the detection of ovarian cancer. The CA125 value of 35 U/mL, in this patient population, has a sensitivity and specificity of 47% and 95%, respectively, for the detection of ovarian cancer. The approximate area under the ROC curve for ovarian cancer patients versus screening patients was 0.817 and 0.753 for YKL-40 and CA125, respectively. The approximate area under the ROC curve for ovarian cancer patients versus benign gynecologic processes was 0.857 and 0.761 for YKL-40 and CA125, respectively.

Serum YKL-40 Levels in Stage I and II Patients and Disease-Free Survival

Thirty-one patients in the study had stage I and II tumors, and 29 of these patients had long-term follow-up (99 months; range, 33 to 125 months) and were assessable for disease recurrence. Of these 29 patients, 10 (34%) had recurrence of disease. All ten patients that had disease recurrence had preoperative YKL-40 values that were more than 80 ng/mL. In fact, of the 29 patients, 14 had YKL-40 values more than 80 ng/mL, and 10 had disease recurrence (71%). Of fifteen patients with preoperative YKL-40 values less than 80 ng/mL, all remain in remission. Moreover, of the 14 patients with YKL-40 values more than 80 ng/mL, nine have died of ovarian cancer; and of the 15 patients with YKL-40 values less than 80 ng/mL, none have died of ovarian cancer.
Figure 4 depicts the recurrence-free survival curves for patients with stage I and II ovarian cancer. The two curves illustrate that patients who had preoperative serum YKL-40 values that were more than 80 ng/mL had shorter disease-free survival than patients with YKL-40 values less than 80 ng/mL (P = .034 based on the log-rank test for log-YKL-40 as a continuous-valued predictor). Neither CA125 nor CA15-3 was predictive of recurrence in this patient population (data not shown).

Discussion

YKL-40 is a glycoprotein in the chitinase protein family, although it does not have chitinase activity. The function of YKL-40 is unknown; however, the pattern of tissue expression suggests that YKL-40 could function in tissue remodeling and angiogenesis. YKL-40 has been evaluated previously as a serum marker for non-neoplastic as well as neoplastic disease.
In this study, serum levels of YKL-40 were assessed in a group of normal individuals, in patients at high risk for developing ovarian cancer, and in patients with ovarian cancer. Serum YKL-40 levels distinguished normal individuals and high-risk patients from ovarian cancer patients very reliably (P < .0001). Strikingly, preoperative serum YKL-40 levels were elevated in 72% of all ovarian cancer patients tested. By contrast, 46% and 26% of the patients had elevated serum CA125 and CA15-3 values. YKL-40 was a significantly better predictor of ovarian cancer than CA125 and CA15-3 in this study population (P < .008).
Early-stage ovarian cancer is frequently not associated with elevated serum markers, including CA125. Importantly, in this study, YKL-40 was abnormal in stage I and II ovarian cancer patients 65% of the time, compared with 35% and 13% for CA125 and CA15-3, respectively (P = .039). YKL-40 also reliably predicts advanced-stage and recurrent ovarian cancer. YKL-40 levels increased with increasing stage and disease burden of ovarian cancer (P < .005; Fig 2A). Importantly, YKL-40 predicted the presence of ovarian cancer regardless of tumor grade, histologic subtype, or patient age, while neither CA125 nor CA15-3 was reliably elevated in mucinous, endometroid, or clear-cell tumors. Thus, YKL-40 deserves more testing as a serum marker for ovarian cancer.
The ROC curves presented reveal that YKL-40 has a sensitivity of 72% and a specificity of 90% for the prediction of ovarian cancer in this study. The sensitivity and specificity values for CA125 for the detection of ovarian cancer were 46% and 95%, respectively. Serum YKL-40 assessment might be combined with other tumor markers, such as CA125 and CA15-3, to increase the sensitivity and specificity of ovarian cancer detection. Using all three markers, it was possible to detect 74% of early-stage tumors.
Since YKL-40 has been described as a serum marker for arthritic conditions, as well as glioblastoma, breast, and colon cancer, serum YKL-40 measurements might be less reliable for the detection of ovarian cancer in patients with these diagnoses. In fact, some of the elevated YKL-40 values in the normal population might have been owing to confounding conditions, such as arthritis. The YKL-40 serum marker would likely be more effective if patients with osteo- and rheumatoid arthritis were not tested with this marker for the detection of ovarian cancer because of the risk of false-positive results in this group. Importantly, our evaluation of the high-risk screening individuals revealed that YKL-40 values were reliably low in this group, comparable to those of the normal individuals. This suggests that YKL-40 testing can be performed to screen high-risk patients for the development of ovarian cancer. Intriguingly, in this study we identified three patients with elevated YKL-40 values from 8 to 18 months before the diagnosis of malignancy or radiographic abnormality.
Finally, analysis of the disease-free survival of stage I and II patients was evaluated in this study based on the degree of elevation of YKL-40. Patients with an elevation of serum YKL-40 more than 80 ng/mL were significantly more likely to have recurrence of disease (P = .034) compared with patients with values less than 80 ng/mL. Furthermore, 64% of patients with YKL-40 values more than 80 ng/mL died of disease, while none of the patients with lower preoperative YKL-40 values died. Of note, preoperative levels of CA125 and CA15-3 in these patients did not correlate with a poor outcome (data not shown). Thus, YKL-40 can identify early-stage patients who are at high risk for recurrence and disease-related death. This could well influence treatment decisions. A recent study has illustrated the potential utility of YKL-40 as a prognostic marker in breast cancer patients.31
In this study, we have identified YKL-40 as a potential serum marker for the detection of early-stage ovarian cancer. In this small series, YKL-40 appears to outperform the established tumor markers CA125 and CA15-3. The YKL-40 ELISA also has the advantage of being commercially available, easily reproducible, and inexpensive. These factors represent an advantage, at present, over serum proteomics,32 which has recently generated much excitement for the detection of ovarian cancer. Furthermore, we have shown that preoperative YKL-40 elevation in early-stage ovarian cancer may have prognostic implications.
Additional prospective studies of YKL-40 as a serum marker are warranted to define the performance of the marker throughout the history of an ovarian cancer patient, including preoperative, postoperative, chemotherapy, and tumor-monitoring periods.

Authors' Disclosures of Potential Conflicts of Interest

The authors indicated no potential conflicts of interest.
Fig 1. Preoperative YKL-40 (ng/mL) values for ovarian cancer (ca) patients versus control groups (P < .0001). Box plots comparing YKL-40 values in preoperative ovarian cancer patients with those of normal individuals, individuals with benign gynecologic (GYN) processes, and high-risk screening individuals with or without prior breast cancer (BrCa). Values for one ovarian cancer patient and two normal controls with rheumatoid arthritis are indicated by “x”; these values were eliminated from statistical comparisons.
Fig 2. Preoperative YKL-40 (ng/mL) values for ovarian cancer patients considering (A) stage of disease, (B) tumor grade, (C) histologic diagnosis, and (D) age. All plots are on logarithmic scale. Vertical lines show log-scale mean (marked by cross-bar) ± two standard deviations. Plot for age includes a curve representing a cubic spline trend estimate. NS, not significant.
Fig 3. Receiver operating characteristics curves of YKL-40 and CA125 in ovarian cancer (ca) compared with (A) screening patients, P = .199; (B) normal controls, P = .045; and (C) benign gynecologic (GYN) disease patients, P = .121.
Fig 4. Time to recurrence of stage I and II ovarian cancer patients considering preoperative elevation level of serum YKL-40 (P = .034). Two groups were compared: patients with YKL-40 values less than 80 ng/mL and more than 80 ng/mL.
Table 1. Serum YKL-40, CA125, and CA15-3 Values for Normal, High-Risk Screening Population, Benign Gynecologic Disorder, and Ovarian Cancer Patients
 NormalHigh-Risk Screening (no cancer)High-Risk Screening (prior cancer)Benign Gynecologic DiseaseOvarian Cancer
No. of Patients4619423350
YKL value, ng/mL     
    Mean33.5*38*44.4*39.6*121.8
    Median283644.53894
    Range15-1669-695-1335-6717-517
    One SD13.712.119.115.399.2
    Mean + two SDs61628270320.2
CA125 value, U/mL     
    Mean13.410.816.921272
    Median11.57.015.512.530
    Range4-315-265-335-2745-2460
CA15-3 value, U/mL     
    Mean17NPNPNP41.5
    Median15.5NPNPNP23
    Range7-34NPNPNP5-470
Abbreviations: SD, standard deviation; NP, not performed.
*
P value not significant for prior cancer versus no cancer, for prior cancer versus normal, or for benign gynecologic disease versus normal based on log (YKL-40).
P < .001 for screening patients with prior breast cancer versus screening patients with no cancer; P = .015 for screening patients with prior breast cancer versus normal; based on log (CA125).
Table 2. Epithelial Ovarian Cancer Patient Demographics
 No. of Patients%
Total50 
Age, years  
    Median59 
    Range31-81 
Primary disease site  
    Ovarian4692
    Fallopian tube36
    Peritoneal12
Stage  
    I2040
    II1122
    III/IV1122
    Recurrent816
Grade  
    148
    2918
    33774
Histologic diagnosis  
    Serous2244
    Endometroid816
    Mucinous510
    Clear cell510
    Other1020
Patient status  
    NED3060
    Recurred1734
    Living3774
    Deceased1020
    Unknown36
Abbreviation: NED, no evidence of disease.
Table 3. Preoperative Serum YKL-40, CA125, and CA15-3 Values for Ovarian Cancer Patients
 Elevated YKL-40 (≥ 62 ng/mL) Elevated CA125 (> 35 U/mL) P Value* for YKL-40 v CA125Elevated CA15-3 (> 35 U/mL) 
 No. of Patients%No. of Patients% No. of Patients%
Total36/507223/5046.00813/5026
Primary disease site       
    Ovarian32/467020/4643.01311/4624
    Fallopian tube3/31002/367.991/333
    Peritoneal1/11001/1100.991/1100
Stage       
    I/II20/316511/3135.0394/3113
    III/IV10/11917/1164.256/1155
    Recurrent6/8755/863.993/838
Tumor grade       
    13/4752/450.990/40
    29/91004/944.0632/922
    324/376517/3746.10911/3730
Histologic diagnosis       
    Serous16/227312/2255.21910/2245
    Endometroid6/8753/838.1250/80
    Mucinous4/5800/50.1250/50
    Clear cell1/5201/520.990/50
    Other9/10907/1070.993/1030
*
Based on McNemar's test for discordant pairs.
Mixed histology (eg, clear cell/papillary serous, adenocarcinoma, poorly differentiated, papillary serous/endometroid).

Acknowledgments

We gratefully acknowledge the assistance of Dr Jeff Boyd and the MSKCC GYN Tumor Bank.
Supported by the Society of Memorial Sloan-Kettering Cancer Prevention Control and Population Science Program Grant, American Society of Clinical Oncology Young Investigator (YIA) and Career Development (CDA) Awards (J.D.), Epithelial Ovarian Program Project (NIH: PO1 CA52477-12), and The Eileen Genet Fund.
Jakob Dupont and Meena K. Tanwar contributed equally to the manuscript.
Authors' disclosures of potential conflicts of interest are found at the end of this article.

References

1.
Greenlee RT, Hill-Harmon MB, Murray T, et al: Cancer Statistics, 2001. CA Cancer J Clin 51::15,2001-36,
2.
Hoskins WJ, Perez C, Young RC (eds): Principles and Practices of Gynecologic Oncology, 2nd ed. Philadelphia, JB Lippincott, 1997
3.
Bast RC Jr: Status of tumor markers in ovarian cancer screening. J Clin Oncol 21::200,2003-205, (suppl 10)
4.
Bast RC Jr, Klug TL, St. John E: A radioimmunoassay using a monoclonal antibody to monitor the course of epithelial ovarian cancer. N Engl J Med 309::883,1983-887,
5.
Rustin GJ, Nelstrop A, Stilwell J, et al: Savings obtained by CA125 measurements during therapy for ovarian cancer. The North Thames Ovary Group. Eur J Cancer 28::79,1992-82,
6.
Rustin GJ, Nelstrop AE, McClean P, et al: Defining response of ovarian carcinoma in initial chemotherapy according to serum CA125. J Clin Oncol 14::1545,1996-1551,
7.
Eisenhauer EA, Vermorken JB, van Glabbeke M: Predictors of response to subsequent chemotherapy in platinum pretreated ovarian cancer. A multivariant analysis of 704 patients. Ann Oncol 8::963,1997-968,
8.
Bast RC Jr, Xu FJ, Yu YH, et al: CA125: The past and the future. Int J Biol Markers 13::179,1998-187,
9.
Scambia G, Benedetti Panici P, Baiocchi G, et al: CA15-3 as a tumor marker in gynecologic malignancies. Gynecol Oncol 30::265,1988-273,
10.
Tuxen MK, Soletormos G, Dombernowsky P: Tumor markers in the management of patients with ovarian cancer. Cancer Treat Rev 21::215,1995-245,
11.
Diamandis EP, Yousef GM, Luo LY, et al: The new human kallikrein gene family: Implications in carcinogenesis. Trends Endocrinol Metab 11::54,2000-60,
12.
Yousef GM, Diamandis EP: The new human tissue kallikrein gene family: Structure, function and association to disease. Endocr Rev 22::184,2001-204,
13.
Clements J, Hooper J, Dong Y, et al: The expanded human kallikrein (KLK) gene family: Genomic organization, tissue specific expression, and potential function. Biol Chem 382::5,2001-14,
14.
Diamandis EP, Scorilas A, Fracchioli S, et al: Human kallikrein 6 (hK6): A potential serum biomarker for diagnosis and prognosis of ovarian carcinoma. J Clin Oncol 21::1035,2003-1043,
15.
Luo LY, Katsaros D, Scorilas A, et al: The serum concentration of human kallikrein 10 represents a novel biomarker for ovarian cancer diagnosis and prognosis. Cancer Res 63::807,2003-811,
16.
Yu JX, Chao L, Chao J: Prostasin is a novel human serine protease from seminal fluid. Purification, tissue distribution, and localization in prostate gland. J Biol Chem 269::18843,1994-18848,
17.
Xu Y, Shen Z, Wiper DW, et al: Lysophosphatidic acid as a potential biomarker for ovarian and other gynecological cancers. JAMA 280::719,1998-723,
18.
Renkema GH, Boot RG, Muijsers AO, et al: Purification and characterization of human chitotriosidase, a novel member of the chitinase family of proteins. J Biol Chem 270::2198,1995-2202,
19.
Renkema GH, Boot RG, Au FL, et al: Chitotriosidase, a chitinase, and the 39-kDa human cartilage glycoprotein, a chitin-binding lectin, are homologues of family 18 glycosyl hydrolases secreted by human macrophages. Eur J Biochem 251::504,1998-509,
20.
Kurose K, Zhou XP, Araki T, et al: Frequent loss of PTEN expression is linked to elevated phosphorylated Akt levels, but not associated with p27 and cyclin D1 expression, in primary epithelial ovarian carcinomas. Am J Pathol 158::2097,2001-2106,
21.
Matsumoto T, Tsurumoto T: Serum YKL-40 levels in rheumatoid arthritis: Correlations between clinical and laboratory parameters. Clin Exp Rheumatol 19::655,2001-660,
22.
Register TC, Carlson CS, Adams MR: Serum YKL-40 is associated with osteoarthritis and atherosclerosis in non-human primates. Clin Chem 47::2159,2001-2161,
23.
Cintin C, Johansen JS, Christensen IJ, et al: Serum YKL-40 and colorectal cancer. Br J Cancer 79::1494,1999-1499,
24.
Tanwar MK, Gilbert M, Holland EC: Gene expression microarray analysis reveals YKL-40 to be a potential serum marker for malignant character in human glioma. Cancer Res 62::4364,2002-4368,
25.
Johansen JS, Cintin C, Jorgensen M, et al: Serum YKL-40: A new potential marker of prognosis and location of metastases of patients with recurrent breast cancer. Eur J Cancer 31A::1437,1995-1442,
26.
Dehn H, Hogdall EV, Johansen JS, et al: Plasma YKL-40, as a prognostic tumor marker in recurrent ovarian cancer. Acta Obstet Gynecol Scand 82::287,2003-293,
27.
Henrissat B, Bairoch A: New families in the classification of glycosyl hydrolases based on amino acid sequence similarities. Biochem J 293::781,1993-788,
28.
Johansen JS, Williamson MK, Rice JS, et al: Identification of proteins secreted by human osteoblastic cells in culture. J Bone Miner Res 7::501,1992-512,
29.
Hensley ML, Robson ME, Kauff ND, et al: Pre- and postmenopausal high-risk women undergoing screening for ovarian cancer: Anxiety, risk perceptions, and quality of life. Gynecol Oncol 89::440,2003-446,
30.
DeLong ER, DeLong DM, Clarke-Pearson DL: Comparing the areas under two or more correlated receiver operating characteristic curves: A nonparametric approach. Biometrics 44::837,1988-845,
31.
Jensen BV, Johansen JS, Price PA: High levels of serum HER2/neu and YKL-40 independently reflect aggressiveness in metastatic breast cancer. Clin Cancer Res 9::4423,2003-4434,
32.
Petricoin EF, Ardekani AM, Hitt BA, et al: Use of proteomic patterns in serum to identify ovarian cancer. Lancet 359::572,2002-577,

Information & Authors

Information

Published In

Journal of Clinical Oncology
Pages: 3330 - 3339
PubMed: 15310777

History

Published in print: August 15, 2004
Published online: September 21, 2016

Permissions

Request permissions for this article.

Authors

Affiliations

Jakob Dupont
From the Developmental Chemotherapy Service, Department of Medicine; Clinical Genetics Service; Departments of Surgery, Neurology, and Cell Biology; Department of Epidemiology and Biostatistics; and Department of Laboratory Science, Memorial Sloan-Kettering Cancer Center; and Joan and Sanford I. Weill Medical College of Cornell University, New York, NY
Meena K. Tanwar
From the Developmental Chemotherapy Service, Department of Medicine; Clinical Genetics Service; Departments of Surgery, Neurology, and Cell Biology; Department of Epidemiology and Biostatistics; and Department of Laboratory Science, Memorial Sloan-Kettering Cancer Center; and Joan and Sanford I. Weill Medical College of Cornell University, New York, NY
Howard T. Thaler
From the Developmental Chemotherapy Service, Department of Medicine; Clinical Genetics Service; Departments of Surgery, Neurology, and Cell Biology; Department of Epidemiology and Biostatistics; and Department of Laboratory Science, Memorial Sloan-Kettering Cancer Center; and Joan and Sanford I. Weill Medical College of Cornell University, New York, NY
Martin Fleisher
From the Developmental Chemotherapy Service, Department of Medicine; Clinical Genetics Service; Departments of Surgery, Neurology, and Cell Biology; Department of Epidemiology and Biostatistics; and Department of Laboratory Science, Memorial Sloan-Kettering Cancer Center; and Joan and Sanford I. Weill Medical College of Cornell University, New York, NY
Noah Kauff
From the Developmental Chemotherapy Service, Department of Medicine; Clinical Genetics Service; Departments of Surgery, Neurology, and Cell Biology; Department of Epidemiology and Biostatistics; and Department of Laboratory Science, Memorial Sloan-Kettering Cancer Center; and Joan and Sanford I. Weill Medical College of Cornell University, New York, NY
Martee L. Hensley
From the Developmental Chemotherapy Service, Department of Medicine; Clinical Genetics Service; Departments of Surgery, Neurology, and Cell Biology; Department of Epidemiology and Biostatistics; and Department of Laboratory Science, Memorial Sloan-Kettering Cancer Center; and Joan and Sanford I. Weill Medical College of Cornell University, New York, NY
Paul Sabbatini
From the Developmental Chemotherapy Service, Department of Medicine; Clinical Genetics Service; Departments of Surgery, Neurology, and Cell Biology; Department of Epidemiology and Biostatistics; and Department of Laboratory Science, Memorial Sloan-Kettering Cancer Center; and Joan and Sanford I. Weill Medical College of Cornell University, New York, NY
Sibyl Anderson
From the Developmental Chemotherapy Service, Department of Medicine; Clinical Genetics Service; Departments of Surgery, Neurology, and Cell Biology; Department of Epidemiology and Biostatistics; and Department of Laboratory Science, Memorial Sloan-Kettering Cancer Center; and Joan and Sanford I. Weill Medical College of Cornell University, New York, NY
Carol Aghajanian
From the Developmental Chemotherapy Service, Department of Medicine; Clinical Genetics Service; Departments of Surgery, Neurology, and Cell Biology; Department of Epidemiology and Biostatistics; and Department of Laboratory Science, Memorial Sloan-Kettering Cancer Center; and Joan and Sanford I. Weill Medical College of Cornell University, New York, NY
Eric C. Holland
From the Developmental Chemotherapy Service, Department of Medicine; Clinical Genetics Service; Departments of Surgery, Neurology, and Cell Biology; Department of Epidemiology and Biostatistics; and Department of Laboratory Science, Memorial Sloan-Kettering Cancer Center; and Joan and Sanford I. Weill Medical College of Cornell University, New York, NY
David R. Spriggs
From the Developmental Chemotherapy Service, Department of Medicine; Clinical Genetics Service; Departments of Surgery, Neurology, and Cell Biology; Department of Epidemiology and Biostatistics; and Department of Laboratory Science, Memorial Sloan-Kettering Cancer Center; and Joan and Sanford I. Weill Medical College of Cornell University, New York, NY

Notes

Address reprint requests to Jakob Dupont, MD, Memorial Sloan-Kettering Cancer Center, Howard 905, 1275 York Ave, New York, NY 10021; e-mail: [email protected]

Metrics & Citations

Metrics

Altmetric

Citations

Article Citation

Download Citation

If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.

For more information or tips please see 'Downloading to a citation manager' in the Help menu.

Format





Download article citation data for:
Jakob Dupont, Meena K. Tanwar, Howard T. Thaler, Martin Fleisher, Noah Kauff, Martee L. Hensley, Paul Sabbatini, Sibyl Anderson, Carol Aghajanian, Eric C. Holland, David R. Spriggs
Journal of Clinical Oncology 2004 22:16, 3330-3339

View Options

View options

PDF

View PDF

Get Access

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Personal login Institutional Login

Purchase Options

Purchase this article to get full access to it.

Purchase this Article

Subscribe

Subscribe to this Journal
Renew Your Subscription
Become a Member

Media

Figures

Other

Tables

Share

Share

Share article link

Share