|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 1999 American Society for Clinical Oncology Quantitative Polymerase Chain Reaction for the Detection of Micrometastases in Patients With Breast CancerFrom the Cancer Research Campaign Laboratories, Department of Cancer Medicine, Imperial College School of Medicine; Department of Surgery, Charing Cross Hospital; and Department of Haematology, Imperial College School of Medicine, Hammersmith Hospital, London, United Kingdom. Address reprint requests to Martin J. Slade, MD, Cancer Research Campaign Laboratories, Department of Cancer Medicine, Imperial College School of Medicine, St. Dunstan's Rd, London W6 8RP, United Kingdom; email m.slade{at}cxwms.ac.uk
PURPOSE: Previous reports have indicated that reverse transcriptase polymerase chain reaction (RT-PCR) for cytokeratin 19 (CK-19) may be useful in the management of patients with breast cancer. However, the specificity of this technique is low, principally because of a high rate of false-positive results. To improve the specificity of this assay, we developed a quantitative RT-PCR methodology that enables an estimate to be made of the number of CK-19 transcripts in blood and bone marrow samples. PATIENTS AND METHODS: We examined 45 peripheral-blood samples and 30 bone marrow samples from patients with a variety of nonneoplastic conditions using nested RT-PCR for CK-19. We also examined bone marrow and peripheral-blood samples from 23 patients with primary breast cancer and peripheral-blood samples from 37 patients with metastatic breast cancer. The number of CK-19 transcripts was estimated in positive specimens by competitive PCR and normalized to the number of ABL transcripts as an internal control for the quality and quantity of cDNA. RT-PCR results were compared with the numbers of CK-19–positive cells detected by immunocytochemistry. RESULTS: Analysis of samples from patients without cancer enabled us to define an upper limit for the background ratio of CK-19 to ABL transcripts (1:1,000 for blood samples and 1:1,600 for bone marrow samples). Using these figures as cut-off points, elevated CK-19: ABL ratios were detected in peripheral-blood samples of 20 of 37 (54%) patients with metastatic breast cancer and in bone marrow samples of 14 of 23 (61%) patients with primary breast cancer. Only three of 23 (13%) primary breast cancer peripheral-blood samples and none of the control samples were positive by these criteria. Only two of 23 patients (9%) with primary breast cancer showed immunocytochemically detectable cells in the blood; 10 of 23 (43%) showed immunocytochemically detectable cells in the bone marrow. Of 36 patients with metastatic breast cancer, eight (22%) showed positive events. CONCLUSION: Quantitative RT-PCR for CK-19 detects a percentage of patients with breast cancer and may enable the progression or regression of the disease to be monitored.
DEATH FROM CARCINOMA of the breast is principally caused by the presence of distant metastases. More than 95% of patients who present with breast carcinoma will have no evidence of metastatic disease on clinical, radiologic, and biochemical examination.1 The presence of bone marrow metastases has been correlated with early recurrence and shorter overall survival. However, a proportion of patients relapse despite the absence of histologic or immunohistochemical evidence of bone marrow micrometastases after resection of the primary tumor.2 Immunocytochemical methods have been used to detect micrometastases, the occurrence of which is related to other prognostic features of the primary carcinoma (tumor size, presence of vascular invasion, lymph node involvement) and predicts for early recurrence.2-4 Immunocytochemical methods have been estimated to be capable of detecting approximately one cancer cell per 104 to 105 normal bone marrow cells,5,6 whereas measurement of epithelial cell-specific gene transcripts such as cytokeratin 19 (CK-19) by reverse transcriptase polymerase chain reaction (RT-PCR) has been reported by our group and others as being capable of detecting one cancer cell per 106 peripheral-blood mononuclear cells.7 RT-PCR in this context has proven controversial, because the specificity with which malignant cells can be detected depends on the number of amplification cycles and the design of the primers. False-positive results are thought to occur from three sources: (1) amplification of low-level, illegitimately transcribed CK-19 from hematopoietic cells, (2) amplification of CK-19 pseudogenes from contaminating genomic DNA,8 and (3) amplification of CK-19 transcripts from contaminating epithelial cells. On the other hand, false-negative results may occur because of the deficient expression of the marker gene in micrometastatic tumor cells. Furthermore, the absence of reliable quantification by RT-PCR has meant that results are generally expressed as either positive or negative, which makes it difficult to relate the level of RT-PCR–detectable disease to the micrometastatic load as judged by immunocytochemistry. We previously developed a competitive RT-PCR titration assay for the leukemia-specific BCR-ABL fusion gene to quantitate levels of residual disease in chronic myeloid leukemia patients after treatment.9 This assay enables the early detection of relapse after bone marrow transplantation and determination of patient response to interferon-alpha?10,11 Here, we have developed a competitive RT-PCR assay for CK-19 and used it to compare levels of transcripts in patients with different stages of breast cancer with the number of cancer cells detected in both blood and bone marrow using immunocytochemistry.
Study Population Samples of blood and bone marrow from each of the left and right posterior iliac crests were first obtained from 23 unselected patients with primary breast cancer from the Charing Cross Hospital breast cancer unit. All patients had cytologically confirmed primary breast cancer and no evidence of distant metastases on chest radiology and bone and liver scanning. Peripheral-blood samples (20 mL) were also obtained from an additional 37 patients with evidence of distant metastatic disease proven cytologically and histologically. We also obtained peripheral-blood samples from some patients who were having blood taken for a variety of reasons; none of these patients had evidence of carcinoma at any site. Some of these individuals underwent surgery and consented to have bone marrow aspirations (see previous paragraph) performed under general anesthesia. The blood (20 mL) was collected in 10-mL Vacutainers (Becton Dickinson, Cowley, United Kingdom) to which 150 units of preservative-free heparin had been added. To avoid epithelial contamination, the first 10 mL of blood was discarded. For bone marrow aspirates, the skin was incised before the aspirates were taken to minimize the risk of epithelial contamination. Between 2 and 5 mL of bone marrow was aspirated from each side using disposable 15-gauge (1.8 mm) marrow-gauge bone marrow aspirate needles (Rocket Medical, Watford, United Kingdom) into syringes primed with preservative-free heparin. The samples were immediately processed as described in this section (see Preparation of Blood and Bone Marrow Samples). The study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethical Review Board. All patients provided written, informed consent.
Preparation of Blood and Bone Marrow Samples
RT-PCR
Construction of Competitor Template
pCKABL-3 is 4.9 kb in size; therefore, 1 ng consists of 1.94 x 108 double-stranded molecules or 3.88 x 108 single-stranded PCR targets.13 After digestion of 10 µg pCKABL-3 with EcoRI, serial dilutions were prepared in 1 mmol/L Tris pH 8.0, 0.1 mmol/L EDTA, 50 µg/mL Escherichia coli tRNA. Dilutions were made in the range of 107 to 10 targets per 2.5 µL, with steps at every half order of magnitude on a logarithmic scale; ie, 107, 3.2 x 106, 106, 3.2 x 105, and so on.
Competitive PCR Quantification of ABL transcripts as an internal control for the amount and quality of cDNA14 was performed for all samples by a single-step PCR: 2.5 µL of cDNA plus 2.5 µL of competitor dilution were added to 20 µL of ABL mix (ABL mix = 12.5 mmol/L of Tris pH 8.3, 2.2 mmol/L of MgCl2, 62.5 mmol/L of KCl, 0.625 µmol/L of primers A2N and A4-, 0.25 mmol/L each of dATP, dCTP, dTTP, and dGTP, and 30 units/mL of Taq polymerase). Bands were visualized at 385 bp for the ABL gene and at 486 bp for the competitor. Competitive RT-PCR results were expressed as the ratio of CK-19:ABL for specimens that were positive for CK-19. Samples that were negative for CK-19 were expressed as negative/(the number of ABL transcripts detected in the same volume of cDNA).
Primers
CKRI-5'-cagaatTCCAAAGGACAGCAGAAGCCCCAG-3' CK-A5'-TCCGCCCGCTTTGTGTCCCTCGT-3' CK-B5'-AGCATCCTTCCGGTTCTGCTCG-3' CK-C5'GGCGGGCAACGAGAAGCTAACC-3' CK-D5'-TCCCACTTGGCCCCTCAGCGTA-3' A2RI5'-ctgaattcAAGCCCTTCAGCGGCCA-3' ABL5-5'-CAAGAAtTCTTCCACCTCCATGG-3' A2N5'-CCCAACCTTTTCGTTGCACTGT-3' A4-5'-CGGCTCTCGGAGGAGACGATGA-3' Lower case letters indicate base changes that were introduced to create restriction enzyme recognition sites.
Cell Culture
Immunocytochemistry
Sensitivity Assay
Establishment of the Competitive RT-PCR Assay A competitor plasmid, pCKABL-3, was constructed that contained cDNA inserts derived from both the CK-19 and ABL genes (Fig 1). Small plasmid fragments were cloned into each of these cDNAs between the PCR primer binding sites so that the competitor and CK-19 or ABL amplification products could be readily distinguished after agarose gel electrophoresis (Fig 2).
Primers for CK-19 were designed to maximize the sequence difference between the normal gene and its pseudogene. Nested PCR enabled a product of the expected size and sequence to be reproducibly amplified from 10-6 to 10-7 dilutions of MCF-7 cells in CK-19–negative peripheral-blood leukocytes. No product was obtained after nested amplification of 1 µg of genomic DNA, indicating that the primers did not amplify the pseudogene. To estimate the number of CK-19 transcripts in RT-PCR–positive specimens, serial dilutions of linearised pCKABL-3 were added to fixed amounts of test cDNA and the mixture subject to nested PCR. To improve the clarity of the bands, it was necessary to dilute the first-step reaction before seeding the second step (Fig 3, lanes 1 to 5 and 7 to 11). If the initial number of competitor molecules was much higher than that of the sample fusion gene message, then only the competitor band is visible on the final gel (Fig 3, lanes 10 and 11). Conversely, if significantly fewer molecules were added, then only the sample CK-19 band is visible (Fig 3, lane 7). If the starting reaction contained equal numbers of competitor and target molecules, then the gel shows both bands with the ratio of the fluorescence intensity in proportion to their sizes. For the sample shown in Fig 3, the equivalence point was estimated to be at 3 x 103 competitor molecules added.
To validate the assay, competitive RT-PCR was performed on serial dilutions of MCF-7 cells. The CK-19:ABL ratio reduced by the same factor as the dilution, indicating that the assay is linear for at least five orders of magnitude (Fig 4).
Quantification of CK-19 in Patient and Control Samples
All samples were tested initially by nested RT-PCR for CK-19. The number of CK-19 transcripts was estimated for all positive specimens by competitive PCR. In addition, the number of normal ABL transcripts was quantified for all specimens. Table 2 summarizes the results obtained by competitive PCR in samples of blood and bone marrow.
Of the 45 control peripheral-blood samples and 30 control bone marrow samples, 23 (51%) and 18 (60%) were RT-PCR–positive for CK-19, respectively. No significant difference was found between the median numbers of ABL transcripts in CK-19–positive and CK-19–negative specimens (23,400 v 21,600; P = .013 using a Student's t test for unpaired data), which indicated that the failure to detect CK-19 mRNA in some samples was not due to poor quality of cDNA. The median level of CK-19 transcripts in positive bone marrow and peripheral-blood samples was 27 and 21, respectively. The highest percentage detected from the control samples that were RT-PCR–positive for CK-19 was 0.1% (1 CK-19:1,000 ABL transcripts) in peripheral blood and 0.06% (1:1,600) for bone marrow. Therefore, we subsequently considered all patients' samples with a CK-19:ABL percentage Figure 5A shows an example of bone marrow and peripheral-blood samples from primary breast cancer patients (lanes 3 to 11) and peripheral-blood samples from metastatic (lanes 13 to 15) and control patients (lanes 16 to 17) subjected to nested RT-PCR for CK-19. Figure 5B shows samples 13 to 17 from Fig 5A subjected to quantitative CK-19 PCR and demonstrates low-level expression of this gene. Sample 13 (lanes 1 to 3) indicates the presence of 40 (101.6) transcripts. Sample 14 (lanes 4 to 6) ) indicates the presence of 25 (101.4) transcripts. Samples 15 (lanes 8 to 10) and 16 (lanes 11 to 13) show 10 transcripts, and sample 17 (lanes 14 to 16) shows fewer than 10 transcripts. Figure 5C shows quantitative PCR for ABL for the above samples. Sample 13 (lanes 1 to 3) indicates the presence of 32,000 (104.5) transcripts. Sample 14 (lanes 4 to 6) indicates the presence of 40,000 (104.6) transcripts. Samples 15 (lanes 8 to 10) and 16 (lanes 11 to 13) and sample 17 (lanes 14 to 16) show 10,000 (105) transcripts. Therefore, sample 13 resulted in a CK-19:ABL ratio of 1:800, sample 14 resulted in a ratio of 1:1,600, samples 15 and 16 resulted in a ratio of 1:10,000, and sample 17 resulted in a ratio of less than 1:10,000. Using the cut-off criteria previously described, only sample 13 was deemed to be positive.
Of the 37 peripheral-blood samples of metastatic breast cancer patients, 23 bone marrow samples of primary breast cancer patients, and 23 peripheral-blood samples of primary breast cancer patients, 28 (76%), 19 (83%), and 16 (70%) samples, respectively, were positive for CK-19 by nested RT-PCR. Using the cut-off figures, 20 of 37 (54.0%) metastatic peripheral-blood samples and 14 of 23 (61.0%) primary patient bone marrow samples were considered positive. Only three of 23 (13%) primary patient peripheral-blood samples were considered positive (Table 3 and Fig 6).
Comparison With Immunocytochemistry
Our study shows that it is possible to quantify cytokeratin transcripts using a quantitative method in which a competitor sequence is used in a PCR titration assay. Increasing amounts of competitor are titrated against an unknown, and thus an estimate of the number of transcripts in a given sample can be made. The technique is of particular value in the case of CK-19 transcript amplification, because it is known that if a sufficient number of cycles of nested PCR are performed, it is generally possible to detect CK-19 transcripts in any sample, thus leading to spurious false-positive results. Although estimation of the equivalence point could be improved (for example, with direct densitometry), we have aimed to keep the method as simple as possible and have previously shown that the variation thus obtained is low.9 Cytokeratins, and in particular CK-19, have been used as tissue-specific markers of metastatic disease in tissues that do not normally express them.16,17 CK-19 has been reported previously not to be expressed by lymphoid or hematopoietic cells.16 However, it has been reported that as many as 40% of healthy patients have CK-19 transcripts in the blood,18,19 and that it is present in non–Hodgkin's lymphoma patients20 and in 20% of control subjects' peripheral-blood mononuclear cells.21 Possible explanations for these false-positives include amplification of illegitimate RNA transcripts17,21 and amplification of the processed CK-19 pseudogene from contaminating genomic DNA, because the CK-19 pseudogene is virtually identical to the CK-19 cDNA sequence.22 To eliminate these false-positives, we have developed a quantitative PCR protocol. This enables us to establish a cut-off point so that known control samples that are positive for CK-19 can be identified as false-positives. We also designed primers that spanned the regions of maximal differences between the pseudogene and the legitimate CK-19 cDNA.
To validate the technique, we "spiked" peripheral blood with MCF7 cells: one cell/107 mononuclear cells was reliably detected; however, we were only able to reliably quantitate one cell/2 x 106 mononuclear cells. This compares favorably with our previously described assay17 in which we were able to detect one MCF7 in 106 mononuclear cells. We analyzed 45 control peripheral-blood samples and found 22 to be negative for CK-19 by RT-PCR and all others to have a CK-19:ABL ratio of We compared this technique with the quantification by the use of immunocytochemistry in both blood and bone marrow. The result of this analysis (Table 3 and Fig 6) indicates that there is a good correlation between the two techniques when they are applied to the metastatic blood samples. Only one of the eight samples that was negative by RT-PCR was shown to be positive by immunocytochemistry. When bone marrow samples of primary breast cancer patients were analyzed, there was a 50% correlation between the two techniques. The probable explanation of this is that we are working in many cases at the limits of the assays, and therefore, some variation is expected because of sampling errors. In addition, the samples that were positive by RT-PCR but negative by immunocytochemistry may have been so due to the superior sensitivity of PCR. The samples that were positive by immunocytochemistry but negative by RT-PCR may have been so because of the fact that we were staining for CK-8 and CK-18, as well as CK-19. As in our previous study in which we compared PCR with immunocytochemistry,23 it seems that PCR is a more sensitive technique than immunocytochemistry. In addition, this study confirms that bone marrow is more likely to be positive than peripheral blood in patients with primary breast cancer (14 of 23 bone marrow samples compared with three of 23 peripheral-blood samples by PCR and 10 of 23 bone marrow samples compared with two of 23 peripheral-blood samples by immunocytochemistry). It has been demonstrated that 50% to 80% of breast cancer patients will develop bone marrow metastases,24 so although the figure of 61% positivity for the bone marrow samples is high, it is within the published range. Because the proof of clinical applicability of new assays lies in the clinical correlation with the assay, we have undertaken a study involving monthly peripheral-blood samples from 22 patients with known metastatic breast cancer. The natural history of the circulating malignant cells and the variation in their number with treatment has been investigated by RT-PCR and immunocytochemistry and is the subject of a further article (Smith et al, manuscript in preparation). The practical implications of this study are two-fold: First, it may be possible to use the assay to monitor patients with micrometastases after primary surgery and adjuvant chemotherapy. At present, it seems that conventional immunocytochemistry is unlikely to be sensitive enough and is subject to a greater degree of sampling errors.2 Second, it may be possible to use the assay in the area of metastatic breast cancer, because it is often difficult to assess response to chemotherapy in these patients. Biochemical markers are frequently misleading,25 and radiologic techniques are insensitive. The only major drawback with quantitative RT-PCR is that it is time consuming. However, as the development of real-time automated PCR now becomes available for the monitoring of hematologic malignancies, this problem will be surmountable.26 Further studies with full follow-up will be needed to clarify this issue and to consider the relative value of quantitative PCR in relation to other tests for determining the prognosis of primary breast cancer.
Supported by the Cancer Research Campaign, the North Thames Regional Health Authority, and the Leukaemia Research Fund. We thank Sisters Jackie English and Helen Graham for their help in recruiting and counseling patients.
1. Coombes RC, Powles TJ: Tests for distant metastases in patients with breast cancer. J R Soc Med 73:617-623, 1980[Medline] 2. Mansi JL, Easton D, Berger U, et al: Bone marrow micrometastases in primary breast cancer: Prognostic significance after 6 years' follow-up. Eur J Cancer 27:1552-1555, 1991 3. Coombes RC, Berger U, Mansi J, et al: Prognostic significance of micrometastases in bone marrow in patients with primary breast cancer. NCI Monogr 1:51-53, 1986 4. Diel I, Kaufmann M, Krempien B, et al: Immunocytochemical detection of tumour cells in bone marrow in patients with primary breast cancer. Br J Cancer 62:3A, 1990 (abstr) 5. Ellis G, Ferguson M, Yamanaka E, et al: Monoclonal antibodies for detection of occult carcinoma cells in bone marrow of breast cancer patients. Cancer 63:2509-2514, 1989[Medline]
6.
Osborne MP, Wong GY, Asina S, et al: Sensitivity of immunocytochemical detection of breast cancer cells in human bone marrow. Cancer Res 51:2706-2709, 1991 7. Schoenfeld A, Luqmani Y, Sinnett HD, et al: Keratin 19 mRNA measurement to detect micrometastases in lymph nodes in breast cancer patients. Br J Cancer 74:1639-1642, 1996[Medline]
8.
Zippelius A, Kufer P, Honold G, et al: Limitations of reverse transcriptase polymerase chain reaction analyses for detection of micrometastatic epithelial cancer cells in bone marrow. J Clin Oncol 15:2701-2708, 1997
9.
Cross NCP, Lin F, Chase A, et al: Competitive polymerase chain reaction to estimate the number of BCR-ABL transcripts in chronic myeloid leukemia patients after bone marrow transplantation. Blood 82:1929-1936, 1993
10.
Lin F, van Rhee F, Goldman JM, et al: Kinetics of increasing BCR-ABL transcript numbers in chronic myeloid leukemia patients who relapse after bone marrow transplantation. Blood 87:4473-4478, 1996
11.
Hochhaus A, Lin F, Reiter A, et al: Quantification of residual disease in chronic myelogenous leukemia patients on interferon alfa therapy by competitive polymerase chain reaction. Blood 87:1549-1555, 1996 12. Cross NCP, Hughes TP, Lin F, et al: Minimal residual disease after allogeneic bone marrow transplantation for chronic myeloid leukemia in first chronic phase: Correlation with acute graft-versus-host disease and relapse. Br J Haematol 84:67-74, 1993[Medline] 13. Sambrook J, Fritsch EF, Maniatis T: Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, NY, Cold Spring Harbor Laboratory Press, 1989 14. Hochhaus A, Lin F, Reiter A, et al: Variable numbers of BCR-ABL transcripts persist in CML patients who achieve complete cytogenetic remission with interferon alpha. Br J Haematol 91:126-131, 1995[Medline] 15. Pantel K, Schlimok G, Angstwurm M, et al: Methodological analysis of immunocytochemical screening for disseminated epithelial tumor markers in bone marrow. J Hematother 3:165-173, 1994[Medline] 16. Datta YH, Adams PT, Drobyski WR, et al: Sensitive detection of occult breast cancer by the reverse transcriptase polymerase chain reaction. J Clin Oncol 12:475-482, 1994[Abstract]
17.
Schoenfeld A, Luqmani Y, Smith D, et al: Detection of breast cancer micrometastases in axillary lymph nodes by using polymerase chain reaction. Cancer Res 54:2986-2990, 1994 18. Burchill SA, Bradbury MF, Pittman K, et al: Detection of epithelial cancer cells in peripheral blood by reverse transcriptase polymerase chain reaction. Br J Cancer 71:278-281, 1995[Medline] 19. Gunn J, McCall JL, Yun K, et al: Detection of micrometastases in colorectal cancer patients by K19 and K20 by reverse transcriptase polymerase chain reaction. Lab Invest 75:611-616, 1996[Medline] 20. Traweek ST, Liu J, Battifora H: Keratin gene expression in non-epithelial tissues: Detection with polymerase chain reaction. Am J Pathol 142:1111-1118, 1993[Abstract] 21. Krismann M, Todt B, Schroder J, et al: Low specificity of cytokeratin 19 reverse transcriptase polymerase chain reaction analysis for detection of hematogenous lung cancer dissemination. J Clin Oncol 13:2769-2775, 1995[Abstract] 22. Bader BL, Jahn L, Franke WW: Low level expression of cytokeratins 8, 18 and 19 in vascular smooth muscle cells of human umbilical cord and in cultured cells derived therefrom, with an analysis of the chromosomal locus containing the cytokeratin 19 gene. Eur J Cell Biol 47:300-319, 1988[Medline] 23. Schoenfeld A, Kruger KH, Gomm JJ, et al: The detection of micrometastases in the peripheral blood and bone marrow of patients with breast cancer using immunohistochemistry and polymerase chain reaction for keratin 19. Eur J Cancer 33:854-861, 1997 24. DeVita VT, Hellman S, Rosenberg SA (eds): Cancer: Principles and Practice of Oncology (ed 3). Philadelphia, PA, Lippincott, 1989 25. Seckl MJ, Rustin GJS, Coombes RC: CA125 is not a reliable marker in metastatic breast cancer. Br J Cancer 66:875-876, 1992[Medline] 26. Morgan GJ, Pratt G: Modern molecular diagnostics and the management of haematological malignancies. Clin Lab Haematol 20:135-141, 1998[Medline] Submitted July 23, 1998; accepted November 3, 1998.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 1999 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
5 has been described previously.
these levels as negative and all samples with a CK-19:ABL percentage greater than these levels as positive. Borderline samples were repeated and the mean of the two results was taken. 
) indicates patients with immunocytochemically detected micrometastases; (
) indicates patients whose disease is stable or in remission.
