|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2005.04.3810 on May 8 2006 © 2006 American Society of Clinical Oncology. Quantitative Fluoroestradiol Positron Emission Tomography Imaging Predicts Response to Endocrine Treatment in Breast Cancer
From the Division of Medical Oncology, University of Washington, Seattle Cancer Care Alliance; Division of Nuclear Medicine; and the Department of Biochemistry, University of Washington School of Medicine, Seattle, WA Address reprint requests to Hannah M. Linden, MD, Division of Medical Oncology, University of Washington, Seattle Cancer Care Alliance, 825 Eastlake Ave East, Seattle, WA 98109; e-mail: hmlinden{at}u.washington.edu
PURPOSE: In breast cancer, [18F]fluoroestradiol (FES) positron emission tomography (PET) correlates with estrogen receptors (ER) expression and predicts response to tamoxifen. We tested the ability of FES-PET imaging to predict response to salvage hormonal treatment in heavily pretreated metastatic breast cancer patients, predominantly treated with aromatase inhibitors. PATIENTS AND METHODS: Initial FES uptake measurements in 47 patients with ER-positive tumors were correlated with subsequent tumor response to 6 months of hormonal treatment. Most patients had bone dominant disease and prior tamoxifen exposure. Response was compared to initial FES-PET uptake, measured qualitatively and quantitatively using standardized uptake value (SUV) and estradiol-binding flux. RESULTS: Eleven of 47 patients (23%) had an objective response. While no patients with absent FES uptake had a response to treatment, the association between qualitative FES-PET results and response was not significant (P = .14). However, quantitative FES uptake and response were significantly associated; zero of 15 patients with initial SUV less than 1.5 responded to hormonal therapy, compared with 11 of 32 patients (34%) with SUV higher than 1.5 (P < .01). In the subset of patients whose tumors did not overexpress HER2/neu, 11 of 24 patients (46%) with SUV higher than 1.5 responded. CONCLUSION: Quantitative FES-PET can predict response to hormonal therapy and may help guide treatment selection. Treatment selection using quantitative FES-PET in our patient series would have increased the rate of response from 23% to 34% overall, and from 29% to 46% in the subset of patients lacking HER2/neu overexpression. A multi-institutional collaborative trial would permit definitive assessment of the value of FES-PET for therapeutic decision making.
Nearly two thirds of breast cancer patients have hormone receptor–bearing tumors, and the incidence of hormone receptor–positive disease appears to be increasing.1 Measurement of hormone receptor expression (estrogen receptor [ER] and progesterone receptor [PR]) at the time of primary diagnosis is the standard of clinical care.2 Knowledge of hormone receptor expression is essential for selection of appropriate therapy.3,4,5 However, treatment selection for metastatic breast cancer poses several challenges. In many metastatic sites, biopsy is associated with significant morbidity or sampling error. ER expression may be heterogeneous; ER expression at one site does not guarantee expression at all sites.6,7 Bone, a common metastatic site,8 poses a particular challenge, due to sampling error and to decalcification of potentially resulting in loss of epitopes, including ER. In patients with a clinical and radiographic presentation consistent with metastatic recurrence, clinicians may be unwilling to withhold endocrine therapy for a negative biopsy. Taken together, the potential for misidentification of patients appropriate for endocrine therapy in the metastatic setting is significant. While ER expression predicts response in 30% to 77% of patients with a new diagnosis of breast cancer, objective response is more typically 7% to 21% in patients with recurrent disease and prior breast cancer treatment.8-13 A predictive assay capable of estimating ER function at all sites of metastatic disease would be highly valuable. Positron emission tomography (PET) ER imaging using [18F]fluoroestradiol (FES) poses an attractive possibility to meet this clinical need. Previous work evaluating the ER binding, radiation dosimetry, blood clearance, and protein interactions of FES in women with hormone receptor positive breast cancer has shown that uptake correlates with in vitro expression assay.14,15 Furthermore, FES-PET evaluates the heterogeneity of ER function in metastatic disease.16,17 Prior studies have shown that the level of FES uptake predicts the response of advanced breast cancer to hormonal therapy, largely tamoxifen.18 We hypothesized that the level of FES uptake, as an in vivo indicator of ER function would predict response to endocrine therapy. We measured PET-FES uptake in a heavily pretreated patient population with metastatic breast cancer that was treated with endocrine therapy. Herein we report the predictive value of qualitative and of quantitative FES-PET in a series of patients with metastatic breast cancer who were treated with hormonal therapy—principally aromatase inhibitors, current first-line therapy for metastatic ER-positive breast cancer.11-13
Patients Forty-seven patients with recurrent or metastastic breast cancer with ER-positive tumors were evaluated. Tumors were evaluated for ER, PR, and HER2/neu expression as described previously.19 Patients participated in one of three ongoing studies at our center evaluating1: the heterogeneity of FES uptake in patients with advanced breast cancer2; the correlation of FES uptake to in vitro assay of ER3; and the change in FES uptake during hormonal therapy. The current analysis includes all patients participating in any of these studies who underwent endocrine treatment without cytotoxic chemotherapy close to the time of FES-PET, and who had an ER-positive primary tumor confirmed by immunohistochemistry and had at least 6 months of follow-up after FES-PET to evaluate response. Patients were required to have at least one site of disease at least 1.5 cm maximal dimension amenable to imaging. Patients with liver metastasis were excluded since FES is hepatically cleared. One hundred and fourteen patients were enrolled in ongoing FES imaging studies. Exclusion from response analysis were due to the following: six patients had excision of measurable disease before 6 month follow-up; 14 patients had insufficient follow-up; and 47 patients received chemotherapy (in addition to endocrine therapy) immediately following FES (as planned by their clinician before the FES study). Concomitant bisphosphonate therapy and/or trastuzumab was allowed. Study protocols were approved by the University of Washington's institutional review board (Seattle, WA) and radioactive drug research committee. Each patient provided informed consent. Patients were required to have discontinued tamoxifen for at least 2 months before FES imaging to allow assessment of tumor uptake of the imaging agent without blockade. FES imaging was performed before hormonal treatment, or shortly after the initiation of treatment. Patients were treated with a minimum of 6 months of hormonal therapy in the absence of overt tumor progression. The type of hormonal therapy was at the discretion of the treating physician. Referring physicians and patients were not blinded to FES-PET results; however, FES-PET results were not used to select or direct therapy.
16-
PET Imaging All patients were imaged using the GE Advance positron emission tomograph (GE, Waukesha, WI) in two dimensional high sensitivity mode with 4.25 mm slice thickness.21,22 For quantitative analysis, dynamic imaging from injection to 60 minutes was performed over a 15 cm body region containing the most prominent tumor sites. Thoracic or upper abdominal regions of tumor involvement, if present, were included preferentially in dynamic imaging to enable estimation of the blood clearance curve from the ventricular cavity. Following a 20- to 25-minute transmission study, 20 mL of FES was infused over 2 minutes and imaged using a sequence previously utilized for FDG imaging.23 For dynamic data collection, then a torso survey covering 5 x 15 cm axial fields of view was performed using 5-minute emission and 3-minute postinjection transmission scans per field of view covering neck to pelvis. Attenuation-corrected torso survey images were used for qualitative image interpretation of the presence or absence of ER at each active site of disease. Images were reconstructed as previously described,22 resulting in approximately 10 mm reconstructed spatial resolution. To aid visual interpretation, iterative reconstruction was also performed for the torso survey. FDG-PET images were obtained close to the time of FES-PET to aid in identification of tumor sites for analysis of the FES images as previously described.25 Quantitative analysis of dynamic FES-PET images used regions of interest drawn on up to three of the largest tumor sites using the summed 30-minute to 60-minute dynamic data, as in prior analyses.23 FDG-PET scans and conventional imaging (computed tomography [CT], bone scan, magnetic resonance imaging [MRI]) were used to draw regions of interest. Partial-volume correction was not used because only sites 1.5 cm or greater diameter were included in quantitative analysis. In patients where the thorax was imaged dynamically, regions were placed over the left ventricular cavity to obtain the blood time activity curve.23
FES uptake at tumor sites used the standardized uptake value (SUV) was calculated as:
 t is the average tumor uptake from 30 minutes to 60 minutes after injection (MBq/mL), ID is the injected dose (GBq), and wt is the patient's weight (kg). We also calculated FES flux, which accounts for variable FES blood clearance, based on preliminary studies suggesting that flux correlated with in vitro ER assay better than SUV.16 FES flux was defined by:
Qualitative analysis was performed prospectively by a single experienced observer without knowledge of subsequent response to hormonal therapy. FES images were compared with FDG-PET and CT, MRI, or bone scan where available. The FES-PET was characterized as either1 having FES uptake above background at all sites of known disease within the imaging field or2 having one or more sites of proven disease, visualized by FDG-PET and confirmed by other imaging modalities, but not seen by FES-PET.
Measurement of Response
Data Analysis
Patient Characteristics Patient population characteristics are listed in Tables 1 and 2. This heterogeneous pretreated group of metastatic breast cancer patients reflects the population referred to our tertiary care center. Most patients were female, had bone-dominant disease (bone only or bone plus soft tissue), and were treated with aromatase inhibitors (AIs). The majority of patients (68%) received prior tamoxifen therapy.
Clinical Response In general, all three reviewers agreed on response (36 of 47; 77%),26 but in 10 instances (21%) the majority opinion was used to characterize patient response. In the single instance in which all three reviewers assessments differed, the patient was categorized as having SD. The  score of 0.72 indicates a high level of interobserver agreement.26 Twenty-three percent of patients were classified as responders, 38% had SD, and 38% had PD (Table 3). No patients had a complete response. However, many responding patients had a sustained response beyond the 6-month response assessment.
Qualitative FES Results Thirteen percent of patients (6 of 47) had one or more FES-negative sites, suggesting loss of receptor function in at least one region of metastatic tumor. All patients negative by qualitative FES had no response to treatment; two of six patients (33%) showed SD; and four of six patients (66%) showed PD (Table 4). Of the patients with qualitative evidence of FES uptake at all sites of disease, 11 of 41 patients (27%) responded, 16 of 41 patients (39%) had SD, and 14 of 41 patients (34%) progressed. The trend for qualitative FES positivity to predict response was not significant (P = .14); analysis of a large subset (44 of 47 patients) with FES performed, on average, within 2 weeks of treatment initiation also showed an insignificant trend (P = .17).
Quantitative FES Results Figure 1 provides illustrative images of baseline FES-PET, and baseline and follow-up FDG-PET.
The observed average FES-SUV values for all lesions/patients included in the dynamic scan ranged from 0.7 to 7.8 with a mean of 2.3. FES flux ranged from 0.008 to 0.065 mL/min/g with a mean of 0.025. Figure 2 shows patients categorized by response, with associated FES-SUV and flux. In testing the ability of FES uptake to classify R versus NR, ROC analysis showed Az equal to 0.69 for SUV and 0.76 for flux. The nature of misclassifications, as expected, occurs at the higher end of the range of both SUV and flux values, where even high thresholds have poor specificity for predicting response.
In analyzing the predictive value of FES uptake, we chose to use the highest value below which no patient had an objective response as the threshold for dichotomization, in order to try to identify all patients who might respond to hormonal treatment. This value was 1.5 for SUV and 0.02 mL/min/g for flux. With these thresholds, 15 of 47 patients had FES-SUV below threshold and 14 of 37 had FES-flux below threshold. Due to tumor locations, 10 patients did not have blood clearance curves needed to assess flux. No patient (zero of 15) with an SUV less than 1.5 had an R to salvage endocrine therapy, whereas 11 of 32 patients (34%) with an SUV higher than 1.5 responded (P < .01; Table 5). Similarly, no patients (zero of 14) with a flux less than 0.02 responded to endocrine therapy, whereas 10 of 24 patients (40%) with a higher flux responded (P < .005). Subset analysis of 44 of 47 patients with FES measured within 2 weeks (mean) of treatment initiation showed similar results; 11 of 31 patients (35%) with an SUV higher than 1.5 had an R to endocrine therapy (P = .01), and 10 of 24 patients (42%) with a flux of higher than 0.02 responded (P < .01).
FES Uptake and Response in Tumors Overexpressing HER2/neu There was no difference in FES uptake between patients with HER2-negative tumors and those with HER2 overexpression (SUV P, .47; flux P, .82; Fig 3); however, no patient with HER2/neu overexpression had an objective response to therapy. Half of the HER2/neu-positive patients (n = 5) were treated with trastuzumab, and four of these patients (80%) had SD, while one (20%) had PD. When analysis of the predictive value of FES was confined to the subset of patients without HER2 overexpression, the Az for the ROC curve for SUV (n = 37) was 0.73 and 0.78 for flux (n = 30). In this subgroup, no patients (zero of 13) with an SUV less than 1.5 had a response, whereas 11 of 24 patients (46%) with an SUV higher than 1.5 responded (P < .004). Similarly, no patients (zero of 12) with a flux < 0.02 mL/min/g responded, whereas 10 of 18 (55%) with flux above this value responded (P < .002).
We found that ER functional in vivo analysis is feasible and effective in predicting response to hormonal therapy in accord with prior studies.18 Patients in our series were selected by clinicians for endocrine treatment based on ER expression in the primary tumor and/or metastases. In a given patient with FES uptake, we noted the absence of ER function by FES-PET in 15% of sites of disease identified as FDG avid, similar to prior FES-PET studies.17 This value is slightly lower than in studies based on biopsy and in vitro assay of expression,6,7 possibly due to reduced sampling error from imaging versus biopsy. Qualitative results showed a statistically insignificant trend toward predicting response, likely limited by small series size, and low fraction of patients without FES uptake. Quantitative FES-PET identified a larger fraction of patients with low FES uptake and significantly predicted response to endocrine therapy. Earlier clinical observations support our findings; absence of ER expression predicts poor response, but not all ER expressing tumors respond to hormonal treatment.9,10 Mortimer et al18 have previously shown that FES uptake predicted response of locally advanced or metastatic breast cancer to tamoxifen therapy. Our study confirms their findings in a different population—heavily pretreated patients treated predominantly with aromatase inhibitors. Like Mortimer, we found that an FES-SUV less than 1.5 to 2.0 predicted a lack of response to hormonal therapy. The use of FES-flux to quantify FES uptake resulted in an improvement in the predictive capability of FES-PET. In our patient population, the selected treatment could be considered salvage endocrine therapy. We observed a response rate of 23% to hormonal therapy, similar to other trials of AIs11-13 and salvage endocrine therapy.27 The use of quantitative FES-PET to select patients for salvage endocrine therapy could have increased the response rate to as high as 40% without excluding any patient who had a response. These findings need to be confirmed independently using prospectively defined quantitative thresholds; however, our results suggest utility for FES-PET as a predictive tool to guide breast cancer therapy selection. In our study, no patients whose tumor overexpressed HER2/neu responded to endocrine treatment. HER2/neu overexpression has been associated with resistance to endocrine therapy.28 One possible explanation is that HER2/neu is associated with a lower average ER expression29,30; however, we found no difference in FES uptake for HER2/neu overexpressing tumors versus nonoverexpressing tumors. Another possible explanation is that HER2/neu overexpression indicates estrogen-independent growth, likely through the epidermal growth factor receptor pathway.31 Our findings are consistent with this hypothesis as tumors with HER2/neu overexpression and high FES uptake failed to respond to hormonal therapy. Exclusion of patients with HER2/neu overexpressing tumors increased the predictive power of FES-PET, with response rates as high as 55% in patients with high FES uptake without HER2/neu overexpression. Another potentially confounding variable is the timing of trastuzumab initiation. Three of five trastuzumab-treated patients were previously treated with a trastuzumab containing regimen and continued on trastuzumab following FES. Two of five trastuzumab treated patients began both hormonal and trastuzumab therapy after FES. While low or absent FES uptake identified patients unlikely to have an objective response to endocrine therapy, it did not exclude a clinical benefit, SD. Several factors may contribute to this finding. SD over 6 months includes patients with disease that may progress slowly, independent of treatment, as well as patients with rapidly PD in whom therapy slows but does not abrogate tumor growth. Our study was not designed to distinguish between these two entities. Other limitations of our study include the difficulty of judging response in bone-dominant breast cancer, and hormonal treatment selection (and thus eligibility for our study) based on enrollment in larger imaging trials, and the clinician's selection of the patient for salvage endocrine therapy rather than more rigorous and uniform eligibility criteria. An additional limitation of our study is timing of FES and initiation of therapy. ER antagonists preclude measurement of FES and delays time to imaging. Preliminary data presented by our group32 shows that serial FES measurements change less than 20% in patients on AI therapy. In conclusion, FES-PET provided incrementally predictive value over standard clinical selection criteria in a challenging and heavily pretreated patient population. Future studies of FES-PET as a predictive assay will benefit from concomitant FES-PET imaging trials with hormonal treatment trials, using uniform selection criteria and treatment regimens. Such trials are warranted on the basis of our results and prior studies,18 and should be feasible with cooperative networks to test new diagnostic imaging agents.
The authors indicated no potential conflicts of interest.
We thank the staff of the University of Washington Radiochemistry Section and PET Imaging facility for technical support and the Seattle Cancer Care Alliance Breast Cancer Specialty Clinic for help with patient referrals.
Supported by NIH Grants No. P01CA42045, R01CA72064, S10 RR17229, and the Avon Foundation. Presented in part at the 51st Annual Meeting of the Society of Nuclear Medicine, Philadelphia, PA, June 19-23, 2004. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Li CI, Daling JR, Malone KE: Incidence of invasive breast cancer by hormone receptor status from 1992 to 1998. J Clin Oncol 21:28-34, 2003 2. Carlson RW, Anderson BO, Cox C, et al: Breast Cancer Clinical Practice Guidelines in Oncology (1 Ed.). National Comprehensive Cancer Network, 2004 3. Early Breast Cancer Trialists' Collaborative Group: Tamoxifen for early breast cancer: An overview of the randomised trials. Lancet 351:1451-1467, 1998[CrossRef][Medline] 4. Breast Cancer (PDQ): Treatment Health Professional Version. Bethesda, MD, National Cancer Institute, 2004 5. Osborne CK: Tamoxifen in the treatment of breast cancer. N Engl J Med 339:1609-1618, 1998 6. Spataro V, Price K, Goldhirsch A, et al: Sequential estrogen receptor determinations from primary breast cancer and at relapse: Prognostic and therapeutic relevance. The International Breast Cancer Study Group (formerly Ludwig Group). Ann Oncol 3:733-740, 1992 7. Kuukasjarvi T, Kononen J, Helin H, et al: Loss of estrogen receptor in recurrent breast cancer is associated with poor response to endocrine therapy. J Clin Oncol 14:2584-2589, 1996[Abstract] 8. Briasoulis E, Karavasilis V, Kostadima L, et al: Metastatic breast carcinoma confined to bone: Portrait of a clinical entity. Cancer 101:1524-1528, 2004[CrossRef][Medline] 9. Osborne CK, Yochmowitz MG, Knight WA III, et al: The value of estrogen and progesterone receptors in the treatment of breast cancer. Cancer 46:2884-2888, 1980[CrossRef][Medline] 10. Bloom ND, Tobin EH, Schreibman B, et al: The role of progesterone receptors in the management of advanced breast cancer. Cancer 45:2992-2997, 1980[CrossRef][Medline] 11. Mouridsen H, Gershanovich M, Sun Y, et al: Superior efficacy of letrozole versus tamoxifen as first-line therapy for postmenopausal women with advanced breast cancer: Results of a phase III study of the International Letrozole Breast Cancer Group. J Clin Oncol 19:2596-2606, 2001 12. Nabholtz JM, Buzdar A, Pollak M, et al: Anastrozole is superior to tamoxifen as first-line therapy for advanced breast cancer in postmenopausal women: Results of a North American multicenter randomized trial. Arimidex Study Group. J Clin Oncol 18:3758-3767, 2000 13. Buzdar A, Douma J, Davidson N, et al: Phase III, multicenter, double-blind, randomized study of letrozole, an aromatase inhibitor, for advanced breast cancer versus megestrol acetate. J Clin Oncol 19:3357-3366, 2001 14. Mankoff DA, Peterson LM, Tewson TJ, et al: [18F]fluoroestradiol radiation dosimetry in human PET studies. J Nucl Med 42:679-684, 2001 15. Mintun MA, Welch MJ, Siegel BA, et al: Breast cancer: PET imaging of estrogen receptors. Radiology 169:45-48, 1988 16. Mankoff DA, Peterson LM, Petra P, et al: Factors affecting the level and heterogeneity of uptake of [F-18] fluoroestradiol (FES) in patients with estrogen receptor positive (ER+) breast cancer. J Nucl Med 43:287P, 2002 17. Dehdashti F, Mortimer JE, Siegel BA, et al: Positron tomographic assessment of estrogen receptors in breast cancer: Comparison with FDG-PET and in vitro receptor assays. J Nucl Med 36:1766-1774, 1995 18. Mortimer JE, Dehdashti F, Siegel BA, et al: Metabolic flare: Indicator of hormone responsiveness in advanced breast cancer. J Clin Oncol 19:2797-2803, 2001 19. Agoff SN, Swanson PE, Linden H, et al: Androgen receptor expression in estrogen receptor-negative breast cancer: Immunohistochemical, clinical, and prognostic associations. Am J Clin Pathol 120:725-731, 2003[CrossRef][Medline] 20. Tewson TJ, Mankoff DA, Peterson LM, et al: Interactions of 16alpha-[18F]-fluoroestradiol (FES) with sex steroid binding protein (SBP). Nucl Med Biol 26:905-913, 1999[CrossRef][Medline] 21. DeGrado TR, Turkington TG, Williams JJ, et al: Performance characteristics of a whole-body PET scanner. J Nucl Med 35:1398-1406, 1994 22. Lewellen T, Kohlmyer S, Miyaoka R, et al: Investigation of the Count Rate Performance of the General Electric ADVANCE Positron Emission Tomograph. IEEE Trans Nucl Sci 42:1051-1057, 1995[CrossRef] 23. Mankoff DA, Dunnwald LK, Gralow JR, et al: Blood flow and metabolism in locally advanced breast cancer: Relationship to response to therapy. J Nucl Med 43:500-509, 2002 24. Therasse P, Arbuck SG, Eisenhauer EA, et al: New guidelines to evaluate the response to treatment in solid tumors: European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 92:205-216, 2000 25. Stafford SE, Gralow JR, Schubert EK, et al: Use of serial FDG-PET to measure the response of bone-dominant breast cancer to therapy. Acad Radiol 9:913-921, 2002[CrossRef][Medline] 26. Landis JR KG: The measurement of observer agreement for categorical data. Biometrics 33:159-174, 1977[CrossRef][Medline] 27. Lonning PE, Taylor PD, Anker G, et al: High-dose estrogen treatment in postmenopausal breast cancer patients heavily exposed to endocrine therapy. Breast Cancer Res Treat 67:111-116, 2001[CrossRef][Medline] 28. Ellis MJ, Coop A, Singh B, et al: Letrozole is more effective neoadjuvant endocrine therapy than tamoxifen for ErbB-1- and/or ErbB-2-positive, estrogen receptor- positive primary breast cancer: Evidence from a phase III randomized trial. J Clin Oncol 19:3808-3816, 2001 29. Konecny G, Pauletti G, Pegram M, et al: Quantitative association between HER-2/neu and steroid hormone receptors in hormone receptor-positive primary breast cancer. J Natl Cancer Inst 95:142-153, 2003 30. Ariazi EA, Clark GM, Mertz JE: Estrogen-related receptor alpha and estrogen-related receptor gamma associate with unfavorable and favorable biomarkers, respectively, in human breast cancer. Cancer Res 62:6510-6518, 2002 31. Osborne CK, Shou J, Massarweh S, et al: Crosstalk between estrogen receptor and growth factor receptor pathways as a cause for endocrine therapy resistance in breast cancer. Clin Cancer Res 11:865s–870s, 2005 32. Linden HM, Link J, Stekhova S, et al: Changes in estrogen binding to ER during endocrine therapy measured by serial [F-18]-fluoroestradiol positron emission tomography (FES PET). Presented at the 28th Annual San Antonio Breast Cancer Symposium, San Antonio, TX, December 8-11, 2005 (abstr 5088) Submitted October 6, 2005; accepted March 20, 2006. This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
TOP
ABSTRACT
INTRODUCTION
-[18F]FES Preparation
2 statistic. The ability of quantitative FES-PET (SUV and flux) to classify response (R versus NR) was analyzed using receiver operator characteristic (ROC) analysis. ROC curves were fit using STATA version 8 (STATA corporation, College Station, TX), and area under the fit of the ROC (Az) was used as a measure of diagnostic performance. ROC analysis was also used to help define quantitative thresholds for FES to define ranges of uptake measures to predict response. Association between the dichotomized uptake values (above or below threshold) and response were tested using the 
.
