|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 22, No 2 (January 15), 2004: pp. 277-285 © 2004 American Society of Clinical Oncology. DOI: 10.1200/JCO.2004.04.148 Prospective Multicenter Study of Axillary Nodal Staging by Positron Emission Tomography in Breast Cancer: A Report of the Staging Breast Cancer With PET Study GroupFrom The Johns Hopkins University School of Medicine, Baltimore, MD; University of Michigan School of Medicine, Ann Arbor, MI; Washington University School of Medicine, St Louis, MO; Duke University School of Medicine, Durham, NC; and Brown University of Medicine, Providence, RI Address reprint requests to Richard L. Wahl, MD, Division of Nuclear Medicine, The Johns Hopkins University School of Medicine, Room 3223 JHOC, 601 N Caroline St, Baltimore, MD 21287; e-mail: rwahl{at}jhmi.edu
PURPOSE: To determine the accuracy of positron emission tomography with fluorine-18–labeled 2-fluoro-2-deoxy-D-glucose (FDG-PET) in detecting axillary nodal metastases in women with primary breast cancer. PATIENTS AND METHODS: In this prospective multicenter study, 360 women with newly diagnosed invasive breast cancer underwent FDG-PET. Images were blindly interpreted by three experienced readers for abnormally increased axillary FDG uptake. Imaging results from 308 assessable axillae were compared with axillary node pathology. RESULTS: For detecting axillary nodal metastasis, the mean estimated area under the receiver operator curve for the three readers was 0.74 (range, 0.70 to 0.76). If at least one probably or definitely abnormal axillary focus was considered positive, the mean (and range) sensitivity, specificity, and positive and negative predictive values for PET were 61% (54% to 67%), 80% (79% to 81%), 62% (60% to 64%), and 79% (76% to 81%), respectively. False-negative axillae on PET had significantly smaller and fewer tumor-positive lymph nodes (2.7) than true-positive axillae (5.1; P < .005). Semiquantitative analysis of axillary FDG uptake showed that a nodal standardized uptake value (lean body mass) more than 1.8 had a positive predictive value of 90%, but a sensitivity of only 32%. Finding two or more intense foci of tracer uptake in the axilla was highly predictive of axillary metastasis (78% to 83% positive predictive value), albeit insensitive (27%). CONCLUSION: FDG-PET has moderate accuracy for detecting axillary metastasis but often fails to detect axillae with small and few nodal metastases. Although highly predictive for nodal tumor involvement when multiple intense foci of tracer uptake are identified, FDG-PET is not routinely recommended for axillary staging of patients with newly diagnosed breast cancer.
Breast cancer is the most common visceral malignancy in women and the second most lethal cancer in the United States, with an estimated 203,500 new cases and 39,600 deaths in 2002 [1]. The most reliable prognostic indicator for recurrence and survival at the time breast cancer is initially diagnosed is the presence and extent of metastasis to the axillary lymph nodes [2,3]. Metastasis to the internal mammary lymph nodes has similar prognostic significance, is more common in medially situated breast cancers, and occasionally represents the only site of locoregional metastasis [4]. Medial tumors have increased recurrence rates and higher mortality than lesions located in other areas of the breast [4]. Currently, axillary lymph node dissection is performed in most women with invasive breast cancer, although sentinel lymph node sampling recently has increased in frequency as an alternative procedure [5,6]. Axillary nodal dissection has well-known risks including lymphedema (12%) and upper extremity dysfunction or discomfort in more than half of patients [7]. Sentinel node biopsy causes less morbidity, but consensus regarding the optimal technique remains lacking and false-negative results occur in a variable number of patients in whom the sentinel nodes are not defined. Furthermore, it is clear that sufficient training of the surgeon performing the procedure is critical to its success [8,9]. There also are concerns that upstaging attributable to the sentinel node procedure can occur as a result of aggressive pathologic sampling and identification of small, hitherto unrecognized tumor foci [10]. The internal mammary and supraclavicular lymph nodes usually are not assessed pathologically in the United States because of the more extensive surgery required to sample these nodes, because there has not been a documented survival difference in most trials comparing extended radical mastectomy with radical mastectomy, and because internal mammary status rarely changes adjuvant therapy [11]. Nonetheless, knowledge of the status of these lymph nodes could be important in clinical decision making. Positron emission tomography (PET) is an imaging method that provides a quantitative portrayal of the in vivo biodistribution of a radioactive tracer, such as the glucose analog fluorine-18–labeled 2-fluoro-2-deoxy-D-glucose (FDG) [12]. Increased use of glucose is a characteristic of most cancers and is in part related to over-expression of the Glut-1 glucose transporter in many cancers, including breast cancer [13]. In the last decade, it has been firmly established that most human cancers, including breast cancer, can be imaged using FDG-PET [14,15]. The Center for Medicare and Medicaid Services authorized expanded coverage for PET imaging in breast cancer for selected indications effective October 2002. Early clinical studies showed that essentially all large primary breast cancers and most known nonnodal soft tissue metastases could be detected by FDG-PET [14-16]. More recent studies have shown that smaller primary breast lesions can escape detection by FDG-PET, likely reflecting resolution and sensitivity limitations of the current scanning systems [17-19]. Several reports showed the feasibility of imaging lymph node metastases from breast cancer with PET [14,20-22]. Early results were quite encouraging, with reported sensitivities in large populations of patients of 90% to 100% and specificities of 75% to 100% [20-22]. In a recent single-institution prospective study of 167 women with newly diagnosed breast cancer, FDG-PET had a sensitivity of 94.4%, a specificity of 86.3%, and an accuracy of 89.8% for detecting axillary metastases [23]. False-negative studies occurred only in women with low tumor burdens. These investigators suggested that FDG-PET could replace axillary nodal dissection and sentinel lymph node biopsy, if their results were reproduced in a multicenter study. However, lower sensitivities for PET have been reported, especially in patients with small primary tumors, and in other studies lower specificities have been reported to achieve high sensitivities for axillary nodal metastases [22,24]. Recent data suggest that detection of internal mammary foci consistent with tumor involvement is feasible [25]. If PET could accurately and noninvasively stage the axilla, as well as the currently unevaluated internal mammary and supraclavicular nodes, it could substantially alter the management of many patients with breast cancer. Patients with negative PET studies could be spared axillary dissections or node sampling procedures, whereas those with positive studies might be treated primarily with chemotherapy. Women with isolated internal mammary nodal metastases could potentially be identified and treated more appropriately, possibly with expanded ports for radiation therapy of these nodes [26]. We performed a prospective, multicenter trial to assess the accuracy of FDG-PET as a method for axillary lymph node staging by comparison with the pathologic findings of conventional axillary dissection as the reference standard. The secondary aims of our study were to assess the prognostic value of PET. We report our results for the primary aim.
This was a prospective multi-center investigation of the accuracy of FDG-PET for detecting locoregional nodal disease in women with newly diagnosed invasive primary breast cancer. The three primary performance sites were the University of Michigan (Ann Arbor, MI), Washington University (St Louis, MO), and Duke University (Durham, NC). Data management was performed by the American College of Radiology (Philadelphia, PA), and the biostatistical center for study design and data analysis was sited at Brown University (Providence, RI). In an effort to facilitate patient accrual, two additional clinical imaging sites, North Shore University Hospital (Manhasset, NY), and Columbia University (New York, NY), were established, but each accrued only a small number of patients This study was approved by the institutional review board at each participating site. Written informed consent was obtained from each patient who participated in the study. Patient accural began on July 19, 1996, and ended on July 28, 2000.
Entry Criteria
Imaging Procedure Patients were imaged after fasting for at least 4 hours. FDG was prepared and quality controlled by the routine methods in use at each performance site in a manner consistent with applicable state and federal regulations. Patients were to be injected with 17 ± 3 mCi of FDG in the arm opposite the primary breast cancer or in a foot vein if bilateral tumors were present. All patients had a fasting blood sugar determination before injection of FDG. Patients rested quietly for 40 minutes after tracer injection. Emission data acquisitions, each of 20-minute duration, were performed at two levels beginning at 50 minutes after tracer injection to include the body region extending from the supraclavicular fossae to the upper abdomen. Transmission data acquisitions, each of 10-minute duration, preceded and followed the respective emission acquisitions at the same levels. In nearly all cases, the patient's arms were extended above the head for the entire image-acquisition period. Images were reconstructed by filtered back-projection using a Hann filter with cutoff frequency equal to 0.6 times the Nyquist frequency or an equivalent and were displayed for interpretation in a 128 x 128 matrix on a nuclear medicine workstation. A quality-control committee assessed image quality.
Image Interpretation
Surgery and Pathologic Review Surgical specimens were examined by a designated study pathologist in the patient's originating site. The protocol for pathology review was developed by a pathology review committee and was common across sites. Breast tumor size was measured and tumors were classified according to the American Joint Committee on Cancer (edition 5) staging criteria. Lymph nodes isolated from axillary fat tissue were formalin fixed, paraffin embedded, and stained with hematoxylin and eosin. Depending on size, each lymph node was sectioned into two or three parts, and one or more sections were prepared from each part. The results of the histologic report were used as a reference to evaluate the accuracy of PET. Pathologists were blinded to FDG-PET interpretations. Patients were observed yearly to assess for recurrence. These methods and results will be described in a subsequent report.
Statistical Analysis Diagnostic performance measures were estimated for each reader separately. For estimating sensitivity, specificity, predictive value, and likelihood ratios, reader degree-of-suspicion data were dichotomized as follows: negative was 0, 1, or 2, and positive was 3 or 4. Standard asymptotic CIs were computed for each quantity. The ROC analysis used the full range of the degree-of-suspicion scale. For each reader, empirical ROC curves and fitted curves were developed using the ROCKIT software, and parametric and nonparametric estimates of the corresponding area under the curve were derived [28]. Parametric estimates are used in Results. The average area and the average sensitivity over all three readers were used to assess the effect of patient characteristics. The computation of SE of differences in average areas or average sensitivities took into consideration correlations in test results attributable to common readers and common cases. The numbers of lymph nodes of each qualitative score were compared with pathologic positivity and test performance characteristics were estimated for varying cutoff points. For semiquantitative analysis, the maximum SUV-lean values of axillary lesions were compared with pathologic positivity and test performance characteristics estimated for varying cutoff points.
Patient Sample A total of 360 patients with newly diagnosed breast cancer were entered onto the study. Three patients were subsequently found to have noninvasive cancer and were excluded; six patients had bilateral cancers, resulting in 363 axillary cases. Intervening therapy was given to 33 patients, and five other patients did not have surgery. Of the 325 remaining cases that had surgery, 15 had sentinel node biopsy only and two had technically unusable PET images. Thus, 308 axillary cases were available for the analysis of the accuracy of PET in axillary nodal staging. These axillary cases had adequate imaging and reference standard information to permit the computation of indices of diagnostic performance. The characteristics of the patients in this analysis group are noted in Tables 1 and 2. The planned accrual target of 414 was not achieved because accrual was slower than expected and funding constraints precluded extending the time of accrual. Notable findings include the generally small size of the primary tumors (median tumor size, 15 mm) and the relatively low prevalence of axillary nodal positivity at pathology (35%) and on physical examination (12%). An example of a positive PET image is shown in Figure 1.
Diagnostic Performance Figure 2 demonstrates the estimated ROC curves for the three readers, using the entire data set of 308 axillary cases. The estimated areas under the curve for the readers range from 0.70 to 0.76. The average area for the observers interpreting the images from their own institutions (primary reads) was slightly but not consistently significantly higher than the corresponding average area for the secondary reads. The estimated areas for the primary (institutional) reads for readers 1, 2, and 3 were 0.77, 0.71, and 0.83 respectively. The corresponding areas for the secondary reads were 0.75, 0.77, and 0.67 respectively. Thus, the data did not indicate the institutional reads to be superior to the centralized secondary reads.
The mean sensitivity, specificity, positive predictive value, and negative predictive value across the three readers were 61% (range, 54% to 67%), 80% (range, 79% to 81%), 62% (range, 60% to 64%), and 79% (range, 76% to 81%), respectively (Table 3). Although the study was neither designed nor powered to allow a subgroup analysis by histology, on retrospective analysis there was a significantly lower average sensitivity for the detection of nodal metastases from invasive lobular carcinoma (25%) than those from other tumor types (64%; P < .005; Table 4). There was a generally good level of concurrence in interpretations among the three readers, with a kappa statistic of 0.63; 33% of the readings were identical across all readers, 34.6% were within 1 scoring unit, and 12.2% were within 2 scoring units of one another. However, in 15 axillary cases, at least one reader interpreted the study as normal, when another interpreted it as definitely abnormal. In a paired comparison between the three sets of two readers, one of the two readers scored the test as positive (on the basis of the dichotomized results), whereas another scored it as negative in 14% to 20% of axillary cases, depending on the specific reader pairs, again indicating good interobserver agreement.
No difference in overall test performance (as measured by the average area under the ROC curve) was observed between patients who had undergone excisional biopsies and those who had undergone no biopsy or a fine-needle or core biopsy. The average area appeared to improve slightly for the second half of the study compared with the first (believed to be a learning effect); however, this difference did not achieve statistical significance. The body mass index, location of the lesion, clinical T stage, image quality, performance site (and equipment), and tumor size did not affect diagnostic performance of the test in our study. The number of images of low quality was significantly associated with a body mass index above 26.5, which was the median for the population evaluated (P < .025; data not shown).
PET images were considered positive by qualitative criteria if there was one or more foci of uptake believed to represent lymph nodes scored 3 or greater. The greater the number and confidence score of nodal tracer uptake, the more likely it was that metastases were present in the axilla. For all readers, the specificity of the test could be increased to at least 97% by requiring the observation of two or more nodes and a confidence score of 4 before the study was considered positive. With this strict criterion, the sensitivity of the test decreased to between 27% and 30%. The number of lymph nodes found positive by pathology was strongly correlated with the number of lymph nodes qualitatively scored as 4 (range of r, 0.52 to 0.63) as was the number of nodes scored 2 to 4 (range of r, 0.56 to 0.57). The number of nodes scored as 2 was not significantly correlated with the number of pathologically positive lymph nodes in the axilla for any reader. A weighted index, calculated as the sum of the scores for each node with a score The sensitivity of PET in the subgroup of patients with only one tumor-involved lymph node at pathology was compared with that in the group of patients with more than one involved lymph node to evaluate the influence of tumor burden on test performance. In patients with only one tumor-involved lymph node, the average sensitivity over the three readers was lower (46%) than in those with more than one tumor-involved node (64%; P = .005). This observation indicates PET to be less sensitive in patients with more limited tumor nodal involvement.
Quantitative Analysis
Physical examination was not a sensitive method for detecting metastatic disease to the axilla (20%). The specificity of physical examination was 93%, and the positive and negative predictive values were 61% and 68%, respectively. The pathologic characteristics of lymph nodes detected and those not detected by PET are shown in Table 6. The mean diameter of the largest lymph node in false-negative axillary cases was 11.5 mm, whereas that for true-positive axillary cases was 15.6 mm. For all readers, the number of tumor-positive nodes (mean, 5.0 v 2.7) and the number of nodes more than 5 mm (mean, 3.2 v 1.4) were significantly higher in the true-positive than in the false-negative groups (P < .003). There was no significant difference in the frequency of lymph nodes with extracapsular invasion when comparing the true-positive and the false-negative axillae (data not shown).
We have shown that PET results were strongly correlated with the status of the axillary lymph nodes, but the results of PET were not equivalent to those of axillary dissection. When PET showed multiple intense lesions in the axilla or when a semiquantitative estimate of glucose metabolism (SUV-lean) was greater than 1.8, there was a high positive predictive value for tumor involvement in the axilla. However, these highly specific interpretation criteria are associated with low sensitivities of less than 30%. Our prospective study shows lower sensitivity of FDG-PET than most (though not all) published reports to date [20-24]. Our results are similar to those of Schirrmeister et al [29], who reported sensitivity and specificity of 79% and 92%, respectively, for FDG-PET in axillary staging in 117 patients. A recent report in 31 women showed the sensitivity of FDG-PET to be only 43%, but with high specificity of 93% [30]. In general, with diagnostic imaging, there is a trade-off between sensitivity and specificity. This was observed by Crippa et al [31] in their 1998 study of PET in breast cancer staging. Such a relationship was also shown in two studies where high sensitivities (100% and 95%) but low specificities (75% and 66%, respectively) were seen [21,24]. These high published sensitivities, even when associated with somewhat lower specificities, are difficult to reconcile with the known performance characteristics of modern PET scanners for the detection of small tumor foci, but could be due, in part, to differences in body habitus or tumor burden among patient populations. We noted lower-quality studies more frequently in our larger patients. Our results with PET show a lower sensitivity than has been reported in the largest single-institution study [23]. The mean size of the primary tumors in that study was 2.2 cm and the prevalence of axillary metastases was 72 of 167 (43%) [23]. The corresponding values in our study were lower (17.9 mm and 34.5%, respectively), suggesting that the patients in the study by Greco et al [23] had more advanced disease, thus potentially resulting in larger, more easily detected nodal metastases. It is also possible the patients in the study from Italy were smaller than our patients, a factor that could potentially influence detectability of lesions. Only approximately 6% of Italian women are reported to be obese [32]. It is possible that use of newer reconstruction algorithms, such as algebraic reconstruction methods, or the use of segmented or computed tomography–based attenuation correction algorithms, could improve our accuracy. Failure to detect small-volume metastatic disease in axillary nodes, as in our study, where FDG-PET was falsely negative in patients with the fewest and smallest nodal metastases, is not surprising with current whole-body PET imaging devices [18,19]. Primary breast cancers, have high glucose metabolism but do not have SUVs as high as those of untreated primary lung cancers [33,34]. Both the size of a lesion and its intensity of tracer uptake are important determinants for lesion detectability with PET. Phantom studies performed by Raylman et al [19] have shown that, with current whole-body PET scanners and the typical soft-tissue background activity levels in women, it is improbable to detect lesions reliably that are smaller than 5 mm in diameter and have the expected FDG uptake of primary breast cancers. Although the number of patients with lobular breast carcinoma was small (n = 30), we observed a low sensitivity for PET in this subgroup. Lobular primary breast cancers have been reported to be less FDG avid than are other types of breast cancer [17]. Our results suggest this to be the case for axillary metastases from lobular carcinoma, as well. Our data clearly demonstrate that false-negative axillary cases with FDG-PET have fewer and generally smaller tumor-involved nodes than do true-positive axillary cases. The average number of tumor-involved lymph nodes in the false-negative axillary cases was 2.7 v 5.1 in true-positive axillary cases. We also demonstrated lower average sensitivity in axillae with only one tumor-positive lymph node than in those with more than one tumor-positive node at pathology. These findings are consistent with those from studies assessing FDG-PET in melanoma in which smaller nodal metastases (< 5 mm) are less frequently detected than larger nodal metastases [35]. Although we did not expect PET to achieve a sensitivity equal to that of axillary dissection, we expected it to have high specificity. The specificity of PET in our study was relatively high, at approximately 80% for our qualitative cutoff. The frequency of false-positive results depends in part on where one sets the imaging threshold cutoff. False-positive results potentially could be reduced in frequency by using quantitative SUV-lean analysis, or by requiring that a larger number of nodes with intense FDG uptake be seen to score a study as positive by qualitative criteria, but at a trade-off to lower sensitivity. However, it was possible in our population to establish thresholds based on the number of nodes and the SUV-lean such that the specificity would exceed 98%, but with reduced sensitivities (Fig 3). High-SUV lesions were more common in patients with axillae considered abnormal or suggestive of positive status on physical examination, although 63% of the axillary cases with high-SUV foci in the axilla had normal findings on axillary palpation. Although PET was less sensitive and less specific than axillary dissection for characterizing the axilla, the PET results were highly correlated with the number of axillary nodes involved by tumor. PET can identify many of the patients with multiple involved axillary nodes and those patients in whom the axillary nodes are enlarged and the soft tissues affected. These patients would be at highest risk of recurrence if tumor were left in the axilla [36]. In addition, there is some evidence that patients with a low tumor burden have a low rate of local recurrence if no axillary dissection is performed [37]. It is likely that the detection by PET of distant metastases and of additional lesions in other nodal groups not assessed by axillary dissection, and the ability of PET to identify patients with larger, multiple, and presumably more ominous axillary metastases, will have prognostic significance. Whether the prognostic value of PET is comparable to that of axillary dissection remains to be ascertained by the follow-up study of our patients. Clearly, PET may have a role in axillary imaging, despite its limitations. It may be rational to consider it for staging of patients who will have neoadjuvant therapy without axillary dissection or sentinel node sampling. In this setting, multiple foci of intense nodal uptake can be seen quite commonly, and, when present, have a high positive predictive value for metastatic tumor to the axilla. PET may also be useful to identify patients who should proceed directly to axillary lymph node dissection, or others in whom sentinel node sampling will be predicted to be sufficient. PET may also have a role in assessing patients with medially or superiorly situated breast cancers that may drain preferentially or exclusively to internal mammary or supraclavicular nodes. Finally, it may be relevant to explore whether patients who have negative results to PET and receive chemotherapy can avoid axillary dissection and sentinel node sampling completely. In summary, this prospective study performed with modern PET scanners demonstrates that FDG-PET is less sensitive and less accurate than axillary lymph node dissection, and currently is not a suitable substitute for axillary nodal histologic evaluation in most patients with breast cancer. However, it is likely that the performance of PET for breast cancer staging will improve with expected technical developments in PET technology, including better imaging devices with higher resolution than current scanners, PET–computed tomography scanners, better tracers, and better image reconstruction algorithms.
The following investigators and research coordinators constituted the Staging Breast Cancer with PET Study Group: University of Michigan School of Medicine, Ann Arbor—R. L. Wahl, A. Chang, J Greenson, P. Saran, K. Zasadny; Washington University School of Medicine, St. Louis, MO—B.A. Siegel, P. Barton, P.D. Cutler, F. Dehdashti, G.M.Doherty, J. Frye, V.M. Hermann, P.A. Humphrey, H. Kammerer, H.M. Maluf, B.S. Monsees, G.W. Philpott, D.M. Radford, D. Trask; Duke University School of Medicine—R.E. Coleman, R. Bentley, P. Kornguth, R. Haithcock, G. Leight, M.S. Soo, T. Turkington; Brown University School of Medicine, Providence, RI—C. Gatsonis, M.-H. Chen, P. Reiss, B. Herman, M. Kim, N. Clement, L. Hanna; North Shore University Hospital, Manhasset, NY—I. Zanzi, A. Belakhlef, B. Babchyck, E. Busch, D. Galgano, M. Kemeny, E. Sellman, A. Yee; Columbia University School of Medicine, New York, NY—R L. Van Heertum, B.A. Ditkoff, R. A. Fawwaz, H. Hibshoosh, P Roseman, L Shriberg; and American College of Radiology, Philadelphia, PA—C. Olson, B. Harrison, K. Parkhurst, S. Sabina, R. Sole, J. Stetz, D. Strybuc.
The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Owns stock (not including shares held through a public mutual fund): Richard L. Wahl, CTI Molecular Imaging; R. Edward Coleman, CTI Molecular Imaging. Acted as a consultant within the last 2 years: Barry A. Siegel, Radiology Corporation of America; R. Edward Coleman, Radiology Corporation of America, General Electric Medical Systems; Richard L. Wahl, Berlex, Millenium, Nilton, Cornixa. Performed contract work within the last 2 years: Richard L. Wahl, General Electric Medical Systems; Barry A. Siegel, Endocyte. Received more than $2,000 a year from a company for either of the last 2 years: Richard L. Wahl, General Electric Medical Systems, Cardinal Health, Glaxo Smith Kline; R. Edward Coleman, Radiology Corporation of America, Synco, General Electric Medical Systems.
Supported by RO1CA66560, awarded by the National Cancer Institute, National Institutes of Health. Authors' disclosures of potential conflicts of interest are found at the end of this article.
1. Jemal A, Thomas A, Murray T, et al: Cancer statistics 2002. CA Cancer J Clin 52:23-47, 2002 2. Cody HS III: Current surgical management of breast cancer. Curr Opin Obstet Gynecol 14:45-52, 2002[CrossRef][Medline] 3. Childberg FW, Lohe F: Basic principles and value of lymphadenectomy in breast carcinoma. Chirurg 67:771-778, 1996[CrossRef][Medline]
4. Lokisch H, Jackson Jones A, et al: Relationship between tumor location and relapse in 6,781 women with early invasive breast cancer. J Clin Oncol 18:2828-2835, 2000 5. Giuliano AE, Kirgan DM, Guenther JM, et al: Lymphatic mapping and sentinel lymphadenectomy for breast cancer. Ann Surg 220:391-398, 1994[Medline]
6. Krag D, Weaver D, Ashikaga T, et al: The sentinel node in breast cancer: A multicenter validation study. N Engl J Med 339:941-946, 1998
7. Kwan W, Jackson J, Weir LM, et al: Chronic arm morbidity after curative breast cancer treatment: Prevalence and impact on quality of life. J Clin Oncol 20:4242-4248, 2002
8. Jakub JW, Pendas S, Reintgen DS: Current status of sentinel lymph node mapping and biopsy: Facts and controversies. Oncologist 8:59-68, 2003 9. Chao C, McMasters K: The current status of sentinel lymph node biopsy for breast cancer. Adv Surg 36:167-192, 2002[Medline] 10. Gerber B, Krause A, Mueller H, et al: Immunohistochemically detected lymph node metastases in breast cancer and their correlation with prognostic factors. Breast J 3:106-111, 1997 11. Klauber-DeMore N, Bevilacqua JL, Van Zee KJ, et al: Comprehensive review of the management of internal mammary lymph node metastases in breast cancer. J Am Coll Surg 193:547-555, 2001[CrossRef][Medline] 12. Wahl RL, Buchanan JW (eds): Principles and Practice of Positron Emission Tomography. Philadelphia, PA, Lippincott Williams and Wilkins, 2002, pp 1-422 13. Brown RS, Wahl RL: Overexpression of Glut-1 glucose transporter in human breast cancer: An immunohistochemical study. Cancer 72:2979-2985, 1993[CrossRef][Medline]
14. Wahl RL, Cody RL, Hutchins GD, et al: Primary and metastatic breast carcinoma: Initial clinical evaluation with PET with the radiolabeled glucose analogue 2-[F-18]-fluoro-2- deoxy-D-glucose. Radiology 179:765-770, 1991 15. Gambhir SS, Czernin J, Schwimmer J, et al: A tabulated summary of the FDG PET literature. J Nucl Med 42:1S-93S, 2001 (5 suppl) 16. Tse NY, Hoh CK, Hawkins RA, et al: The application of positron emission tomographic imaging with fluorodeoxyglucose to the evaluation of breast disease. Ann Surg 216:27-34, 1992[Medline]
17. Avril N, Rose CA, Schelling M, et al: Breast imaging with positron emission tomography and fluorine-18 fluorodeoxyglucose: Use and limitations. J Clin Oncol 18:3495-3502, 2000 18. Budinger TF: PET instrumentation: What are the limits?. Semin Nucl Med 28:247-267, 1998[CrossRef][Medline] 19. Raylman RR, Kison PV, Wahl RL: Capabilities of two- and three-dimensional FDG-PET for detecting small lesions and lymph nodes in the upper torso: A dynamic phantom study. Eur J Nucl Med 26:39-45, 1999[CrossRef][Medline]
20. Adler LP, Crowe JP, al-Kaisi NK, et al: Evaluation of breast masses and axillary lymph nodes with [F-18] 2-deoxy-2-fluoro-D-glucose PET. Radiology 187:743-750, 1993 21. Utech CI, Young CS, Winter PF: Prospective evaluation of fluorine-18 fluorodeoxyglucose positron emission tomography in breast cancer for staging of the axilla related to surgery and immunocytochemistry. Eur J Nucl Med 23:1588-1593, 1996[CrossRef][Medline]
22. Avril N, Dose J, Janicke F, et al: Assessment of axillary lymph node involvement in breast cancer patients with positron emission tomography using radiolabeled 2-(fluorine-18)-fluoro-2-deoxy-D-glucose. J Natl Cancer Inst 88:1204-1209, 1996
23. Greco M, Crippa F, Agresti R, et al: Axillary lymph node staging in breast cancer by 2-fluoro-2-deoxy-D-glucose-positron emission tomography: Clinical evaluation and alternative management. J Natl Cancer Inst 93:630-635, 2001
24. Adler LP, Faulhaber PF, Schnur KC, et al: Axillary lymph node metastases: Screening with [F-18]2-deoxy-2-fluoro-D-glucose (FDG) PET. Radiology 203:323-327, 1997
25. Eubank WB, Mankoff DA, Takasugi J, et al: 18Fluorodeoxyglucose positron emission tomography to detect mediastinal or internal mammary metastases in breast cancer. J Clin Oncol 19:3516-3523, 2001 26. Freedman GM, Fowble BL, Nicolaou N, et al: Should internal mammary lymph nodes in breast cancer be a target for the radiation oncologist? Int J Radiat Oncol Biol Phys 46:805-814, 2000[CrossRef][Medline]
27. Sugawara Y, Zasadny KR, Neuhoff AW, et al: Reevaluation of the standardized uptake value for FDG: Variations with body weight and methods for correction. Radiology 213:521-525, 1999
28. Metz CE, Herman BA, Roe CA, et al: Statistical comparison of two ROC-curve estimates obtained from partially-paired data sets. Med Decis Making 18:110-121, 1998 29. Schirrmeister H, Kuhn T, Guhlmann A, et al: Fluorine-18 2-deoxy-2-fluoro-D-glucose PET in the preoperative staging of breast cancer: Comparison with the standard staging procedures. Eur J Nucl Med 28:351-358, 2001[CrossRef][Medline] 30. Guller U, Nitzsche EU, Schirp U, et al: Selective axillary surgery in breast cancer patients based on positron emission tomography with 18F-fluoro-2-deoxy-D-glucose: Not yet!. Breast Cancer Res Treat 71:171-173, 2002[CrossRef][Medline]
31. Crippa F, Agresti R, Seregni E, et al: Prospective evaluation of fluorine-18-FDG PET in presurgical staging of the axilla in breast cancer. J Nucl Med 39:4-8, 1998 32. Pagano R, La Vecchia C: Overweight and obesity in Italy, 1990-91. Int J Obes Relat Metab Disord 18:665-669, 1994[Medline]
33. Brown RS, Leung JY, Kison PV, et al: Glucose transporters and FDG uptake in untreated primary human non-small cell lung cancer. J Nucl Med 40:556-565, 1999
34. Wahl RL, Zasadny K, Helvie M, et al: Metabolic monitoring of breast cancer chemohormonotherapy using positron emission tomography: Initial evaluation. J Clin Oncol 11:2101-2111, 1993
35. Crippa F, Leutner M, Belli F, et al: Which kinds of lymph node metastases can FDG PET detect? A clinical study in melanoma. J Nucl Med 41:1491-1494, 2000 36. Chua B, Ung O, Taylor R, et al: Is there a role for axillary dissection for patients with operable breast cancer in this era of conservatism? J Surg 72:786-792, 2002[CrossRef] 37. Greco M, Agresti R, Cascinelli N, et al: Breast cancer patients treated without axillary surgery: Clinical implications and biologic analysis. Ann Surg 232:1-7, 2000[CrossRef][Medline] Submitted April 22, 2003; accepted October 31, 2003.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2004 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
2 was correlated with the number of pathologically positive lymph nodes in the axilla (range of r, 0.57 to 0.61). There was a consistent underestimation of the number of pathologically involved lymph nodes by PET. In general, the greater the number of lymph nodes visualized and the greater their score, the greater the likelihood that metastatic cancer was present in the axilla. 
