1 Department of Oncology, Cancer Center Karolinska, Radiumhemmet, Karolinska Institute and University Hospital, Stockholm; 2 Department of Surgery, Östra Hospital, Sahlgrenska University Hospital, Gothenburg; 3 Regional Oncological Center, University Hospital, Uppsala; 4 Gyros AB, Uppsala Science Park, Uppsala, Sweden
* Correspondence to: Dr J. Andersson, Department of Oncology and Pathology, Radiumhemmet, Cancer Center Karolinska Karolinska Institute and Hospital, SE-171 76 Stockholm, Sweden. Tel: +46-8-5177-6279; Fax: +46-8-5177-9524; Email: jenny.andersson{at}cck.ki.se
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Patients and methods:: We investigated relapse-free (RFS), breast cancer-corrected (BCCS) and overall survival (OS) related to TP53 status in node-positive breast cancer patients that had received polychemotherapy [cyclophosphamide, methotrexate, 5-fluorouracil (CMF)] and/or endocrine therapy (tamoxifen). Sequence analyses of the whole TP53 coding region was performed in 376 patients operated on for primary breast cancer with axillary lymph node metastases between 1984 and 1989 (median follow-up time 84 months).
Results:: TP53 mutations were found in 105 patients (28%). We found 90 (82%) of the 110 mutations in the more frequently analysed exons 58, while the other 20 (18%) were located in exons 34 and 910, respectively. Univariate analyses showed TP53 to be a significant prognostic factor with regard to RFS, BCCS and OS in patients who received adjuvant CMF.
Conclusions:: TP53 mutations might induce resistance to certain modalities of breast cancer therapy. Sequence-determined TP53 mutation was of negative prognostic value in the total patient population and in the CMF treated patients.
Key words: adjuvant therapy, CMF, p53, sequence-based analysis, tamoxifen, TP53
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
Patients and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Primary systemic therapy
In this cohort of 376 women, 193 patients received systemic adjuvant chemotherapy. The majority (174 patients) received polychemotherapy with intravenous methotrexate and 5-fluorouracil together with oral cyclophosphamide (CMF) over 36 months. Adjuvant endocrine therapy was given to 251 women; 137 (of whom 127 were postmenopausal) received tamoxifen alone at doses of 20 mg/day. Another 46 women (of whom 44 were postmenopausal) received tamoxifen therapy alone, but at doses of 10, 30 or 40 mg/day. Sixty-eight patients (of whom 61 were postmenopausal) were treated with both adjuvant CMF and tamoxifen.
Clinical follow-up
The patients were mostly seen at regular outpatient visits at the different Departments of Surgery in the Gothenburg region and/or at the Department of Oncology at the Sahlgrenska University hospital during the first 5 years after breast cancer diagnosis. Women participating in prospective and randomised clinical studies (141 patients in the International Breast Cancer Study Group VVII) were controlled according to the routines described for each study. Patients treated outside the clinical studies had laboratory and X-ray investigations taken on the basis of clinical signs and symptoms. Deaths were classified as death caused by breast cancer (145 patients), death due to other cause with signs of active breast cancer (seven patients) and death from another cause without signs of breast cancer (44 patients), based on medical records and death certificates.
Tumour material
All operated primary breast cancers were routinely brought, as soon as possible, on ice, to the Department of Pathology, Eastern University Hospital, Gothenburg, and were handled by one pathologist, Dr Johan Säve-Söderberg (deceased before the initiation of this study). Tumour material was removed for routine histopathology and locoregional staging. Additional tumour pieces were removed for receptor analyses and for future use (this material was stored at 70 °C). These were the standard procedures for all breast cancers in the Gothenburg area from 1983 and throughout the whole study period.
Sequence-based analysis of TP53 status
The details of this procedure have been published in detail elsewhere [14]. Since that publication, minor modifications and improvements have been introduced, described in more detail as follows. Total RNA was extracted from frozen tumour samples under stringent conditions (QuickPrepTM Total RNA Extraction Kit; Amersham-Pharmacia Biotech, Uppsala, Sweden). This procedure allowed the exclusion of phenol and chlorophorm. This procedure was followed by an enzymatic synthesis of cDNA by PCR, using four overlapping primer pairs covering the entire coding region of the TP53 gene. Two of 12 primers were altered compared with the original publication; both of them covered fragment 1 [14
]. The used PCR primer for fragment 1 had the following sequence: B-5'-TCC ACG ACG GTG ACA CGC TTC-3'. The sequence primer consisted of the following basepairs: C-5'-GGA GTA CGT GCA AGT CAC A-3'. One of the primers in each primer pair was modified with biotin. Thus, biotin-labelled PCR products were generated, facilitating solid-phase sequencing [15
]. The fluorescence-labelled primers were this time labelled with Cy5, owing to the fact that the ALF express sequencer was used instead of the original ALF version. A manifold-based version of solid phase sequencing was utilised, essentially as described by Lagerkvist et al. [16
]. The sequencing products generated were analysed using an automated laser fluorescence sequencer (ALFexpressTM; Amersham-Pharmacia Biotech). The generated sequence was then compared with the sequence of wild-type TP53 using a prototype software program TP53 SBDecipher, version 1.00. Only nucleotide changes resulting in amino acid shifts were considered as mutations. Each identified mutation was verified by sequencing an entirely new PCR product, from the corresponding cDNA. Further confirmation was obtained by analysis of the neighbouring PCR fragment when the mutation was located in an overlapping segment.
Statistical methods
Survival was estimated using the KaplanMeier method and log-rank tests were performed for comparison of survival between groups. To estimate simultaneous effects by different factors on survival, Cox proportional hazard regression models were used. We included the factors TP53 status, age at diagnosis, tumour size, number of positive lymph nodes (03 versus 4), estrogen (ER) and progesterone (PR) receptor status, S-phase and level of differentiation (low versus medium and high). All factors were dichotomised in the analyses except for tumour size and age at diagnosis, which were analysed as continuous variables. All hazard ratios (HRs) reported in this paper are derived from these estimated models. Survival was measured from date of diagnosis. In breast cancer-corrected survival (BCCS), dates of death from other causes were handled as censored observations in the analyses. In relapse-free survival (RFS), all end points except for recurrences were handled as censored observations.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
TP53 status and survival analyses
We analysed the total patient population using RFS, BCCS and overall survival (OS). The same analyses were performed in three subgroups based on given adjuvant therapy: tamoxifen therapy only (TAM), CMF only (CMF) and CMF together with tamoxifen (CMFTAM), respectively. In all patients and in the CMF treated patients TP53 status was a statistically significant prognostic factor with regard to RFS (P=0.004 and P=0.01 respectively; not shown). TP53 status was also of statistically significant importance when censoring for BCCS in all patients (P <0.0001; not shown) and in the CMF group (P=0.001; not shown). Patients with TP53 mutations had statistically significantly worse OS compared with wild-type TP53 patients (P=0.0005; Figure 1). Worse OS for women with TP53-mutated tumours was also seen in patients treated with CMF (P=0.001; Figure 1).
|
In ER-positive patients (n=286), regardless of given adjuvant therapy, univariate analyses showed TP53 to be a significant prognostic factor with regard to BCCS (P=0.003; not shown) and OS (P=0.009; not shown), but not RFS. In patients with ER-negative tumours (n=82), regardless of given adjuvant therapy, TP53 was of no significant importance for RFS, BCCS or OS.
Cox proportional hazards statistics
Multivariate analyses by means of Cox proportional hazards models of RFS showed age at diagnosis, tumour size and number of positive lymph nodes to be independent prognostic factors in the total study population (Table 2) as well as in the tamoxifen treated patients (not shown). Tumour size was the only independent prognostic factor in the CMF-(Table 2) and in the CMFTAM subgroups (not shown). The estimated HR for tumour size was only marginally higher than 1.0, however. For BCCS in the whole study population, tumour size, number of positive lymph nodes and level of differentiation were found to be independent prognostic factors (not shown). Subgrouping with regard to given adjuvant therapy singled out tumour size as an independent prognostic factor in all three groups (CMF, TAM and CMFTAM), but again with HRs around 1.0. For the TAM patients the number of positive lymph nodes and level of differentiation were also independent (not shown), and for the CMFTAM patients the PR status was also of significance regarding BCCS (not shown). In Cox's analyses of the total study population with reference to OS, tumour size, number of positive lymph nodes and ER status were independent statistically significant prognostic factors (Table 3). Subgrouping according to given adjuvant therapy showed tumour size to be the only statistically significant factor in the CMF group (Table 3), while the number of positive lymph nodes was the only factor of significance in the TAM group (not shown).
|
|
Cox proportional hazards statistics in ER-positive patients, regardless of given adjuvant therapy, confirmed the prognostic value of TP53 with regard to BCCS [HR 1.88; 95% confidence interval (CI) 1.163.05].
Mutation type and location
Multivariate analyses of the entire study population and of the subgroup treated with adjuvant CMF showed that stop codon mutations were of significant importance with regard to BCCS (HR 7.1, 95% CI 1.926; and HR 15, 95% CI 1.4150, respectively) and OS (HR 3.4, 95% CI 1.110; and HR 10, 95% CI 1.198, respectively). These analyses had a small sample size (six stop codons of 110 mutations) and revealed wide CIs. The location of each mutation in different evolutionarily conserved regions (including zinc-finger binding regions) in all and CMF-treated patients was of no statistical importance.
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The mutation frequency was higher in the CMF group compared with the other subgroups (TAM and CMFTAM). CMF patients were younger at the time of diagnosis and presented with more ER-negative tumours, as well as more poorly differentiated tumours. In summary, they had a more aggressive biology. Perhaps a functional TP53 is of greater importance in patients with more aggressive tumours compared with patients with less aggressive tumours.
At the time that our patient cohort received their adjuvant treatment, the recommendation for tamoxifen therapy was based on menopausal status rather than on hormone receptor expression. For this reason, we also analysed the prognostic value of TP53 mutations in tamoxifen-treated ER-positive and -negative tumours, respectively, as the subgroups of CMF, TAM and CMFTAM might be considered to be arbitrary. By means of Cox proportional hazard statistics, TP53 was of no importance with reference to survival analyses in tamoxifen-treated patients, regardless of whether they were ER-positive or -negative. In contrast to our present results, we previously reported a worse survival in patients with TP53 mutations receiving adjuvant tamoxifen and radiotherapy in a subgroup of a small number of node-positive patients (48 patients) [5]. Our present study has the advantage that only a few patients (42/376, 11%; Table 1) received adjuvant radiotherapy, which thus largely avoids an interaction between radiotherapy and TP53 status [20
].
In some reports, TP53 status has been shown to predict poor response to tamoxifen in metastatic disease, as well as being a prognostic factor associated with worse survival after tamoxifen therapy in the adjuvant setting [7, 21
]. In contrast, relatively recent reports have claimed that TP53 status is of no prognostic importance with regard to survival after tamoxifen therapy [22
, 23
]. However, in the latter studies the duration of tamoxifen therapy was short (1 year) or unknown, and in both reports immunohistochemistry was the method used for TP53 evaluation. Several explanations are available for the conflicting results in literature with regard to the predictive value of TP53 status, such as heterogeneity of the patient cohorts and tumours, as well as TP53 methodology and therapeutic regimens. In addition, other studies have been under-powered to track potential differences between patients with tumours containing mutant TP53 and those with wild-type TP53. A review article published in 2000 stated that there is inadequate evidence to support TP53 as a guide when selecting endocrine, chemo- and radiotherapy, which is also our opinion [17
].
This study included 376 population-derived patients operated on for primary breast cancer with positive axillary nodes during 19841989 (median follow-up time 84 months). TP53 mutations were present in 28% of the patients, which is in good concordance with previous reports in patients with lymph node metastases demonstrating 20% to 30% TP53 mutations [35
, 14
]. Slight differences in mutation frequency could be due to different proportions of node-positive patients and detection methodology. Different techniques are available to determine the status of the TP53 gene: immunohistochemistry, cDNA-based sequencing and denaturing gradient gel electrophoresis, among others [14
, 24
26
]. At present, relatively few studies have used more optimal methods, such as cDNA-based sequencing, rather than IHC for TP53 determinations [7
, 14
, 19
, 27
, 28
].
Three years ago, Vogelstein and Kinzler [13] reasoned that TP53 is an Achilles' heel of cancer. Its tumour specificity and wild-type function indicate a possibility of exclusively targeting cancer cells with compounds and strategies that enhance the effect of available drugs. Studies like ours may help in determining the use and feasibility of such TP53 reactivating strategies. Our data imply that women with TP53-mutated breast cancer tumours should be offered treatments other than CMF. For example, Kandolier-Eckersberger et al. [29
] indicate that patients with mutated TP53 might benefit from paclitaxel, while the response to FEC is dependent on wild type TP53, while other data show the reverse [8
]. The predictive therapeutic value of TP53 needs to be further investigated and is presently being studied in the ongoing randomised EORTC p53 study [30
].
![]() |
Acknowledgements |
---|
Received for publication September 3, 2004. Revision received December 20, 2004. Accepted for publication December 22, 2004.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2. Early Breast Cancer Trialists' Collaborative Group. Tamoxifen for early breast cancer: an overview of the randomised trials. Lancet 1998; 351: 14511467.[CrossRef][ISI][Medline]
3. Soussi T. The p53 tumor suppressor gene: from molecular biology to clinical investigation. Ann N Y Acad Sci 2000; 910: 121137. discussion 137139.
4. Overgaard J, Yilmaz M, Guldberg P et al. TP53 mutation is an independent prognostic marker for poor outcome in both node-negative and node-positive breast cancer. Acta Oncol 2000; 39: 327333.[CrossRef][ISI][Medline]
5. Bergh J, Norberg T, Sjogren S et al. Complete sequencing of the p53 gene provides prognostic information in breast cancer patients, particularly in relation to adjuvant systemic therapy and radiotherapy. Nat Med 1995; 1: 10291034.[CrossRef][ISI][Medline]
6. Clahsen PC, van de Velde CJ, Duval C et al. p53 protein accumulation and response to adjuvant chemotherapy in premenopausal women with node-negative early breast cancer. J Clin Oncol 1998; 16: 470479.[Abstract]
7. Berns EM, Foekens JA, Vossen R et al. Complete sequencing of TP53 predicts poor response to systemic therapy of advanced breast cancer. Cancer Res 2000; 60: 21552162.
8. Bertheau P, Plassa F, Espie M et al. Effect of mutated TP53 on response of advanced breast cancers to high-dose chemotherapy. Lancet 2002; 360: 852854.[CrossRef][ISI][Medline]
9. Muss HB, Thor AD, Berry DA et al. c-erbB-2 expression and response to adjuvant therapy in women with node-positive early breast cancer. N Engl J Med 1994; 330: 12601266.
10. Archer SG, Eliopoulos A, Spandidos D et al. Expression of ras p21, p53 and c-erbB-2 in advanced breast cancer and response to first line hormonal therapy. Br J Cancer 1995; 72: 12591266.[ISI][Medline]
11. Elledge RM, Green S, Howes L et al. bcl-2, p53, and response to tamoxifen in estrogen receptor-positive metastatic breast cancer: a Southwest Oncology Group study. J Clin Oncol 1997; 15: 19161922.[Abstract]
12. Makris A, Allred DC, Powles TJ et al. Cytological evaluation of biological prognostic markers from primary breast carcinomas. Breast Cancer Res Treat 1997; 44: 6574.[ISI][Medline]
13. Vogelstein B, Kinzler KW. Achilles' heel of cancer? Nature 2001; 412: 865866.[CrossRef][ISI][Medline]
14. Sjogren S, Inganas M, Norberg T et al. The p53 gene in breast cancer: prognostic value of complementary DNA sequencing versus immunohistochemistry. J Natl Cancer Inst 1996; 88: 173182.
15. Stahl S, Hultman T, Olsson A et al. Solid phase DNA sequencing using the biotinavidin system. Nucleic Acids Res 1988; 16: 30253038.[Abstract]
16. Lagerkvist A, Stewart J, Lagerstrom-Fermer M, Landegren U. Manifold sequencing: efficient processing of large sets of sequencing reactions. Proc Natl Acad Sci USA 1994; 91: 22452249.
17. Hamilton A, Piccart M. The contribution of molecular markers to the prediction of response in the treatment of breast cancer: a review of the literature on HER-2, p53 and BCL-2. Ann Oncol 2000; 11: 647663.[Abstract]
18. Tetu B, Brisson J, Plante V, Bernard P. p53 and c-erbB-2 as markers of resistance to adjuvant chemotherapy in breast cancer. Mod Pathol 1998; 11: 823830.[ISI][Medline]
19. Blaszyk H, Hartmann A, Cunningham JM et al. A prospective trial of midwest breast cancer patients: a p53 gene mutation is the most important predictor of adverse outcome. Int J Cancer 2000; 89: 3238.[CrossRef][ISI][Medline]
20. Jansson T, Inganas M, Sjogren S et al. p53 Status predicts survival in breast cancer patients treated with or without postoperative radiotherapy: a novel hypothesis based on clinical findings. J Clin Oncol 1995; 13: 27452751.[Abstract]
21. Powell BL, Bydder S, Grieu F et al. Prognostic value of TP53 gene mutation in adjuvant treated breast cancer patients. Breast Cancer Res Treat 2001; 69: 6568.[CrossRef][ISI][Medline]
22. Knoop AS, Bentzen SM, Nielsen MM et al. Value of epidermal growth factor receptor, HER2, p53, and steroid receptors in predicting the efficacy of tamoxifen in high-risk postmenopausal breast cancer patients. J Clin Oncol 2001; 19: 33763384.
23. Berry DA, Muss HB, Thor AD et al. HER-2/neu and p53 expression versus tamoxifen resistance in estrogen receptor-positive, node-positive breast cancer. J Clin Oncol 2000; 18: 34713479.
24. Borresen AL, Hovig E, Smith-Sorensen B et al. Constant denaturant gel electrophoresis as a rapid screening technique for p53 mutations. Proc Natl Acad Sci USA 1991; 88: 84058409.
25. Chappuis PO, Estreicher A, Dieterich B et al. Prognostic significance of p53 mutation in breast cancer: frequent detection of non-missense mutations by yeast functional assay. Int J Cancer 1999; 84: 587593.[CrossRef][ISI][Medline]
26. Williams C, Norberg T, Ahmadian A et al. Assessment of sequence-based p53 gene analysis in human breast cancer: messenger RNA in comparison with genomic DNA targets. Clin Chem 1998; 44: 455462.
27. Bergh J. Clinical studies of p53 in treatment and benefit of breast cancer patients. Endocr Relat Cancer 1999; 6: 5159.
28. Soussi T, Beroud C. Assessing TP53 status in human tumours to evaluate clinical outcome. Nat Rev Cancer 2001; 1: 233240.[CrossRef][Medline]
29. Kandioler-Eckersberger D, Ludwig C, Rudas M et al. TP53 mutation and p53 overexpression for prediction of response to neoadjuvant treatment in breast cancer patients. Clin Cancer Res 2000; 6: 5056.
30. Bonnefoi H, Diebold-Berger S, Therasse P et al. Locally advanced/inflammatory breast cancers treated with intensive epirubicin-based neoadjuvant chemotherapy: are there molecular markers in the primary tumour that predict for 5-year clinical outcome? Ann Oncol 2003; 14: 406413.
|