Skip Navigation



Annals of Oncology Advance Access published online on October 22, 2007
Annals of Oncology, doi:10.1093/annonc/mdm358
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
19/1/68    most recent
mdm358v1
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Kim, S. J.
Right arrow Articles by Noguchi, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kim, S. J.
Right arrow Articles by Noguchi, S.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 2007 European Society for Medical Oncology

Determination of the specific activity of CDK1 and CDK2 as a novel prognostic indicator for early breast cancer

S. J. Kim1,§, S. Nakayama2,§, Y. Miyoshi1, T. Taguchi1, Y. Tamaki1, T. Matsushima2, Y. Torikoshi2, S. Tanaka2, T. Yoshida2, H. Ishihara2 and S. Noguchi1,*

1 Department of Breast and Endocrine Surgery, Graduate School of Medicine, Osaka University
2 Sysmex Corporation, Kobe, Japan

* Correspondence to: Dr S. Noguchi, Department of Breast and Endocrine Surgery, Graduate School of Medicine, Osaka University, 2-2-E-10 Yamadaoka, Suita City, Osaka 565-0871, Japan. Tel: +81-6-6879-3772; Fax: +81-6-6879-3779; E-mail: noguchi{at}onsurg.med.osaka-u.ac.jp


   Abstract
Top
 Abstract
introduction
 patients and methods
 results
 discussion
 References
 
Background: We recently established a novel assay for specific activity (SA) of cyclin-dependent kinases (CDKs) using small tumor samples (≥8 mm3). The aim of this study was to investigate the prognostic significance of CDK1SA and CDK2SA in human breast cancer.

Methods: CDK1SA and CDK2SA were determined in 284 breast cancer patients and their prognostic significance was investigated.

Results: Tumors with high CDK1SA and high CDK2SA showed significantly poorer 5-year relapse-free survival than those with low CDK1SA and low CDK2SA, respectively (66.9% vs 84.2% for CDK1SA; 43.6% vs 83.6% for CDK2SA). Moreover, combined analysis of CDK1SA and CDK2SA enabled the classification of breast tumors into high-risk and low-risk groups, where tumors in the high-risk group were strongly associated with unfavorable prognosis (5-year relapse-free survival 69.4% for the high-risk group and 91.5% for the low-risk group). Multivariate analysis showed that the risk determined by combined analysis of CDK1SA and CDK2SA is a significant (hazard ratio 3.09, P < 0.001) prognostic indicator for relapse, especially in node-negative patients (hazard ratio 6.73, P < 0.001).

Conclusion: Determination of CDK1SA and CDK2SA may be useful in the prediction of outcomes in breast cancer patients and has potential for use as a routine laboratory test.

breast cancer, cycline dependent kinase, prognosis


   introduction
Top
 Abstract
 introduction
patients and methods
 results
 discussion
 References
 
It is well established that systemic adjuvant therapy for early breast cancer significantly reduces the risk of recurrence and death regardless of nodal status [1, 2]. However, the fact that approximately two-thirds of node-negative patients can survive without recurrence even without adjuvant therapy indicates that adjuvant therapy is administered to many patients who actually do not need it. To avoid unnecessary treatments, we need new and more powerful prognostic indicators [3, 4].

Recently, molecules involved in cell cycle regulation such as cyclins, cyclin-dependent kinases (CDKs) and CDK inhibitors have been attracting considerable attention as potential prognostic indicators [4–6]. Cyclin E appears to be the most promising of these molecules. High cyclin E expression detected by western blotting has been shown to be strongly associated with unfavorable prognosis, independent of nodal status [5]. However, it is not easy to reproducibly assay total cyclin E or low molecular weight cyclin E expression by western blotting, which does not seem to be suitable for routine laboratory tests.

We have been focusing on CDKs (CDK1 and 2) and investigating their prognostic significance in breast cancers because CDKs play a pivotal role in cell cycle regulation [7, 8]. The CDK expression levels are almost constant but their activities change markedly according to the cell cycle phase. Thus, it is necessary to measure CDK activity itself to accurately evaluate the role of CDKs in cell proliferation. Recently, we succeeded in developing a system that can assay the specific activity (SA) of CDKs using small tissue samples [9]. The aim of this study was to clarify the prognostic implications of CDKSA in breast cancers.


   patients and methods
Top
 Abstract
 introduction
 patients and methods
results
 discussion
 References
 
patients
For this study, 284 patients with primary invasive breast cancer who had undergone mastectomy or breast-conserving surgery between November 1996 and December 2002 were recruited. Of these 284 patients, 162 patients were given hormonal therapy (tamoxifen alone, 124; tamoxifen plus luteinizing hormone-releasing hormone analog, 31; other modalities, 7), 37 patients underwent chemotherapy (cyclophosphamide, methotrexate and 5-fluorouracil [CMF], 16; cyclophosphamide plus epirubicin [CE], 19; other modalities, 2) and 61 patients received chemohormonal therapy (CMF plus tamoxifen, 17; CE plus tamoxifen, 25; other modalities, 19).

The median follow-up period was 56.6 (8–89) months, and the relapse-free survival rate at 5 years after surgery (5yRFS) was 80.9%. Forty-nine patients developed recurrence (liver, 6; lung, 9; bone, 11; soft tissue, 23). Ipsilateral breast recurrences after breast-conserving surgery were not counted as recurrences.

assay for CDKSA
The assay of CDKSA consists of analyses of protein expression and kinase activity, as previously described [9]. In brief, lysates of frozen tissues were prepared with a homogenizer and stored at –80 °C until use. For expression analysis, the lysate was applied to an ImmobiChip (Sysmex, Kobe, Japan). The target protein was detected by sequential reactions with primary antibodies (anti-CDK1, anti-CDK2 or glyceraldehyde-3-phosphate dehydrogenase (GAPDH); Sysmex, Kobe, Japan), biotinylated secondary antibodies (Santa Cruz Biotechnology, Santa Cruz, CA) and fluorescein-labeled streptavidin (Vector, Burlingame, CA). For kinase activity analysis, the CDK1 or CDK2 molecules in the lysate were first captured in a mini-column coupled with anti-CDK1 or anti-CDK2 antibody. Then an in-column kinase reaction and a fluorescein labeling reaction were performed sequentially, and the final reaction mixture was applied to the ImmobiChip. For quantification of both CDK expression and activity, catalytically active recombinant CDK1 or CDK2 (Upstate Biotechnology, Lake Placid, NY) was used as a standard. The CDKSAs were then calculated as kinase activity (U/µL lysate, where 1 U is equivalent to the activity of 1 ng of standard) divided by its corresponding expression (ng/µL lysate). The cut-off values for CDK1SA, CDK2SA and CDK2SA/CDK1SA ratio were defined as the points that gave the best discrimination in RFS. The optimal cut-off points were 100 U/ng for CDK1SA, 800 U/ng for CDK2SA and 5.6 for CDK2SA/CDK1SA. The distribution of breast tumors according to CDK1SA and CDK2SA is shown in Figure 1.


Figure 1
View larger version (20K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. Distribution of breast tumors according to CDK1SA and CDK2SA. Tumors (n = 284) are plotted in two dimensions (logarithmic scales) according to CDK1SA and CDK2SA. Area A includes tumors with high CDK1SA (>100 U/ng) and/or high CDK2SA (>800 U/ng) (n = 37). Area C includes tumors where both CDK1SA and CDK2SA are less than lower measurement limits (n = 33). The remaining tumors are divided into two groups (B1 [n = 85] and B2 [n = 129]) according to the CDK2SA to CDK1SA ratio, with a cut-off at 5.6. Tumors in areas A and B1 are considered to be high-risk for relapse (CDK-based high-risk group) and those in areas B2 and C to be low-risk (CDK-based low-risk group). Tumor without relapse; •, tumor with relapse; Figure 1.

 
assay for human epidermal growth factor receptor type 2 expression
HER2 expression was examined by HercepTest (DakoCytomation, Carpinteria, CA) in 195 patients and by western blotting in 87 patients whose primary tissues were not available for HercepTest. The insoluble membrane fraction of the lysate for CDKSA assay was solubilized by RIPA buffer-supplemented protease inhibitor cocktail (SIGMA-Aldrich, St Louis, MO). The resultant supernatant was electrophoresed followed by transfer to PVDF membrane. After blocking, the membrane was treated with polyclonal anti-HER2 antibody (Upstate Biotechnology, Lake Placid, NY), biotinylated anti-rabbit antibody (Santa Cruz Biotechnology, Santa Cruz, CA) and Alexa-Fluor488-streptavidin (Molecular Probes, Eugene, OR). Fluorescent signal intensities of HER2 were measured and normalized to GAPDH expression. HER2 expression was classified as negative, 1+ or 2+. A high concordance (82%) between score 3+ of HercepTest and 2+ of the western blotting was confirmed (data not shown), and both were defined as HER2-positive.

statistical methods
RFS was calculated with the Kaplan–Meier method, and the differences were assessed with the log-rank test. The Cox proportional hazards model was used for both univariate and multivariate analyses. Test results were considered significant for P ≤ 0.05.


   results
Top
 Abstract
 introduction
 patients and methods
 results
discussion
 References
 
relationship of various clinicopathologic parameters or CDK1/2SA with prognosis
The relationship of various clinicopathologic parameters with 5yRFS is shown in Table 1. Lymph node metastases, high histologic grade, estrogen receptor (ER) negativity, progesterone receptor (PR) negativity and HER2 positivity were significantly associated with poor 5yRFS. With respect to CDKSAs, patients with high CDK1SA and high-CDK2SA tumors showed a significantly lower 5yRFS than those with low CDK1SA and low-CDK2SA tumors, respectively. Moreover, patients with tumors with a high CDK2SA/CDK1SA ratio showed a significantly lower 5yRFS than those with tumors with a low CDK2SA/CDK1SA ratio.


View this table:
[in this window]
[in a new window]

  		Table 1. Association between tumor parameters and 5-year RFS in all patients (n = 284)

 
Next, we studied the relationship of the combination of CDK1SA and CDK2SA with prognosis. Patients with high CDK1SA and/or high-CDK2SA tumors (area A in Figure 1) showed a poor prognosis (5yRFS rate 60%), whereas patients with tumors where both CDK1SA and CDK2SA were less than lower measurement limits (area C in Figure 1) were unlikely to develop recurrent diseases (5yRFS rate 96%). The remaining patients were able to be divided into the high- and low-risk groups according to the CKD2SA/CDK1SA ratio; that is, patients with tumors with a high CKD2SA/CDK1SA ratio (area B1 in Figure 1) were at high risk of relapse (5yRFS rate 73%) and those with a low CKD2SA/CDK1SA ratio (area B2 in Figure 1) were at low risk of relapse (5yRFS rate 91%). Accordingly, using the combination of CDK1SA and CDK2SA, all patients could be classified into a CDK-based low-risk group (area B2 and C in Figure 1) and a CDK-based high-risk group (area A and B1 in Figure 1). Patients in the CDK-based high-risk group showed a significantly lower 5yRFS than those in the CDK-based low-risk group (Table 1 and Figure 2A).


Figure 2
View larger version (16K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2. Relapse-free survival (RFS) rates according to the CDK-based risk and St Gallen's risk categorization. In (A) all, (B) node-positive and (C) node-negative patients, CDK-based high risk was strongly associated with poor prognosis compared to CDK-based low risk. In node-negative patients receiving hormone therapy alone as systemic adjuvant therapy, risk classification according to (D) St. Gallen risk category (2005 version) and (E) histologic grade failed to show a significant difference in RFS. (F) CDK-based risk was able to classify these patients into the high- and the low-risk groups, and their 5-year RFS rates were 74.9% vs 98.4%, respectively (P = 0.0001).

 
The prognostic impacts of various markers were evaluated by univariate and multivariate analyses (Table 2). In the univariate analysis, lymph node status, histologic grade, ER, PR, HER2 and CDK-based risk were significantly associated with relapse. In the multivariate analysis, however, only lymph node status and CDK-based risk had a significant correlation with relapse (hazard ratio 2.22 and 3.09, respectively).


View this table:
[in this window]
[in a new window]

  		Table 2. Univariate and multivariate analyses for relapse in all patients (n = 284)

 
CDK1/2SA and clinicopathologic parameters.
The relationship of CDK-based risk with clinicopathologic parameters was evaluated with the chi-square test. CDK-based high risk showed a significant association with large tumor size (P = 0.035), lymph node involvement (P = 0.046), high histologic grade (P = 0.0008) and PR negativity (P = 0.004), but no significant association with ER (P = 0.362) and HER2 status (P = 0.118).

CDK1/2SA and prognosis according to nodal status.
In both node-negative and node-positive subsets, patients in the CDK-based high-risk group showed a significantly lower 5yRFS than those in the CDK-based low-risk group (node-negative, 72.6% vs 97.8%; node-positive, 61.0% vs 79.0%) (Figure 2B and 2C).

In the node-positive group, univariate analysis showed that the number of metastatic lymph nodes, ER status and CDK-based risk were significantly associated with relapse, whereas multivariate analysis showed only that the number of metastatic lymph nodes and ER status were significant prognostic indicators for relapse (data not shown). In the node-negative group, univariate analysis showed that the CDK-based risk had a significant association with relapse, and that the histologic grade and PR status had a tendency to be associated with relapse. The multivariate analysis demonstrated that only CDK-based risk is a significant independent prognostic indicator (hazard ratio 6.73).

prognostic factors for node-negative patients receiving hormonal therapy alone
Of 178 node-negative patients, 139 (78%) patients received hormone therapy alone as adjuvant therapy, and 14 of these 139 patients developed recurrences. Neither histologic grade nor the St Gallen's criteria [10], widely used as the risk classification especially for node-negative patients, showed a significant association with relapse in these 139 patients (Figure 2D and 2E). However, patients in the CDK-based high-risk group showed a significantly lower 5yRFS than those in the CDK-based low-risk group (74.9% vs 98.4%, P = 0.0001) (Figure 2F).


   discussion
Top
 Abstract
 introduction
 patients and methods
 results
 discussion
References
 
In this study, we applied our novel assay system to breast cancers to find out whether determination of CDK1SA and CDK2SA could be useful for the prediction of patient outcomes. Although a high CDK1SA, a high CDK2SA and a high CDK2SA/CDK1SA ratio were significantly associated with a poor prognosis, the combination of these parameters (the CDK-based risk) has been found to predict patients’ outcomes more accurately than each parameter alone. Multivariate analysis demonstrated that CDK-based risk was a significant prognostic indicator. More importantly, CDK-based risk was a highly significant and independent prognostic indicator for node-negative breast cancers.

The strength of this new indicator, CDK-based risk, is that it classified as many as 61% (109/178) of node-negative patients into the low-risk group where the RFS is extremely good, and the remaining 39% (69/178) into the high-risk group where the RFS is so low as to be equivalent to that seen in patients with one lymph node involvement [11]. This excellent capability for differentiation of the CDK-based risk sharply contrasts with that of St Gallen's risk classification of node-negative breast cancers. The latter categorized only 5% (8/166) of our subjects into the low-risk group, where recurrence was observed in 13% (1/8), and the remaining 95% (158/166) into the intermediate risk group, where recurrence was also observed in 13% (20/158).

We have focused on node-negative patients treated with hormonal therapy alone as systemic adjuvant therapy because this group represents the majority of node-negative cancers and includes some patients with unfavorable prognosis. For these patients, only the CDK-based risk was of significant use for the prediction of their prognosis (5yRFS 74.9% vs 98.4%). These findings seem to indicate that adjuvant hormonal therapy alone is under-treatment for node-negative and hormone receptor-positive patients with tumors belonging to the CDK-based high-risk group, who need chemotherapy in addition to hormonal therapy. By contrast, adjuvant hormonal therapy alone is an appropriate treatment for those in the CDK-based low-risk group. These preliminary findings obtained with a limited number of patients need to be confirmed in a future study including a larger number of patients.

Both CDK1 and CDK2 are considered to play an important role in cell proliferation and are expected to be associated with tumor aggressiveness and a poor prognosis [7, 8, 12, 13]. However, the prognostic impact of CDK1 in breast cancers still remains controversial [13–15]. Interestingly, some recent studies have shown that CDK1 may be required for apoptosis that is independent of the regulation of the cell cycle [16, 17]. Uncontrolled CDK1 activation might work as a brake for cancer cell growth in some tumors. Our present study has shown that a high ratio of CDK2SA to CDK1SA is associated with a poor prognosis and a low ratio is associated with a favorable prognosis. Although the real biological meaning of this ratio is still unclear, implication of CDK1 in apoptosis might partially explain why a low ratio of CDK2SA to CDK1SA is associated with a favorable prognosis. Several in vitro studies to clarify the biological meaning of this ratio are in progress in our laboratory.

Our results have demonstrated that tumors in the CDK-based high-risk group showed a significant association with unfavorable clinicopathologic features, such as high histologic grade, large tumor size, lymph node metastases and negative PR. CDK-based risk has a particularly strong association with histologic grade, suggesting that CDK-based risk may reflect the cell proliferation. It is well established that rapidly proliferating tumors are associated with a malignant potential to metastasize [4]. In fact, various parameters associated with cell growth have been identified as having the capability to serve as prognostic indicators in breast cancers. These parameters include mitotic index, DNA flow cytometry, 3H-thymidine/5-bromo-2'-deoxyuridine uptake and Ki-67 antigen immunohistochemistry [18, 19]. The main problem inherent in these methods is that they are of a subjective nature with significant inter-observer or inter-assay variations, and are thus too difficult to standardize for use in routine laboratory tests. By contrast, determination of CDK1SA and CDK2SA can be accomplished with a well-standardized method ready for use in laboratory tests [9]. Another strength of CDK1SA and CDK2SA assay is that it needs only a very small sample (minimum 8 mm3).

In conclusion, we have shown that CDK-based risk determined by evaluating CDK1SA and CDK2SA is strongly associated with clinical outcome especially for node-negative breast cancer patients. We consider that the CDK-based risk has potential as a new prognostic factor independent of the conventional risk factors, and as a routine laboratory test. However, our results need to be validated in a study with a larger number of patients on a multicenter basis.


   Footnotes
 
§ Both authors contributed equally to this manuscript Back

Received for publication February 27, 2007. Revision received June 12, 2007. Accepted for publication June 14, 2007.


   References
Top
 Abstract
 introduction
 patients and methods
 results
 discussion
 References
 
1. Early Breast Cancer Trialists’ Collaborative Group. Polychemotherapy for early breast cancer: an overview of the randomised trials. Lancet (1998) 352:930–942.[CrossRef][Web of Science][Medline]

2. Early Breast Cancer Trialists’ Collaborative Group. Tamoxifen for early breast cancer: an overview of the randomised trials. Lancet (1998) 351:1451–1467.[CrossRef][Web of Science][Medline]

3. Clark GM. Do we really need prognostic factors for breast cancer? Breast Cancer Res Treat (1994) 30:117–126.[CrossRef][Web of Science][Medline]

4. Colozza M, Azambuja E, Cardoso F, et al. Proliferative markers as prognostic and predictive tools in early breast cancer: where are we now? Ann Oncol (2005) 16:1723–1739.[Abstract/Free Full Text]

5. Keyomarsi K, Tucker SL, Buchholz TA, et al. Cyclin E and survival in patients with breast cancer. N Engl J Med (2002) 347:1566–1575.[Abstract/Free Full Text]

6. Kuhling H, Alm P, Olsson H, et al. Expression of cyclins E, A, and B, and prognosis in lymph node-negative breast cancer. J Pathol (2003) 199:424–431.[CrossRef][Web of Science][Medline]

7. Moroy T, Geisen C, Cyclin E. Int J Bioch Cell Biol (2004) 36:1424–1439.[CrossRef][Web of Science][Medline]

8. Sutherland RL, Musgrove EA. Cyclins and breast cancer. J Mammary Gland Biol Neoplasia (2004) 9:95–104.[CrossRef][Web of Science][Medline]

9. Ishihara H, Yoshida T, Kawasaki Y, et al. A new cancer diagnostic system based on a CDK profiling technology. Biochim Biophys Acta (2005) 1741:226–233.[Medline]

10. Goldhirsch A, Glick JH, Gelber RD, et al. Meeting highlights: international expert consensus on the primary therapy of early breast cancer 2005. Ann Oncol (2005) 16:1569–1583.[Abstract/Free Full Text]

11. Saez RA, McGuire WL, Clark GM. Prognostic factors in breast cancer. Semin Surg Oncol (1989) 5:102–110.[Medline]

12. Lee M-H, Yang H-Y. Regulators of G1 cyclin-dependent kinases and cancers. Cancer Metast Rev (2003) 22:435–449.[CrossRef][Web of Science][Medline]

13. Kourea HP, Koutras AK, Scopa CD, et al. Expression of the cell cycle regulatory proteins p34cdc2, p21waf1, and p53 in node negative invasive ductal breast carcinoma. J Clin Pathol (2003) 56:328–335.

14. Winters ZE, Hunt NC, Bradburn MJ, et al. Subcellular localisation of cyclin B, Cdc2 and p21WAF1/CIP1 in breast cancer: association with prognosis. Eur J Cancer (2001) 37:2405–2412.[CrossRef][Web of Science][Medline]

15. Umemura S, Komaki K, Noguchi S, et al. Prognostic factors for node-negative breast cancers: results of a study program by the Japanese Breast Cancer Society. Breast Cancer (1998) 5:243–249.[Medline]

16. Castedo M, Perfettini J-L, Roumier T, et al. Cyclin-dependent kinase-1: linking apoptosis to cell cycle and mitotic catastrophe. Cell Death Differ (2002) 9:1287–1293.[CrossRef][Medline]

17. Golsteyn RM. Cdk1 and Cdk2 complexes (cyclin dependent kinases) in apoptosis: a role beyond the cell cycle. Cancer Lett (2005) 217:129–138.[CrossRef][Web of Science][Medline]

18. Daidone MG, Silvestrini R. Prognostic and predictive role of proliferation indices in adjuvant therapy of breast cancer. J Natl Cancer Inst Monogr (2001) 30:27–35.[Abstract/Free Full Text]

19. Michels J-J, Marnay J, Delozier T, et al. Proliferative activity in primary breast carcinomas is a salient prognostic factor. Cancer (2004) 100:455–464.[CrossRef][Web of Science][Medline]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
19/1/68    most recent
mdm358v1
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Kim, S. J.
Right arrow Articles by Noguchi, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kim, S. J.
Right arrow Articles by Noguchi, S.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?