|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2006.05.7406 on November 6 2006 © 2006 American Society of Clinical Oncology. Monitoring of Early Response to Neoadjuvant Chemotherapy in Stage II and III Breast Cancer by [18F]Fluorodeoxyglucose Positron Emission Tomography
From the Nuclear Medicine, Pathology, Radiophysics, Statistics, Radiology, and Medical Oncology Units, René Gauducheau Cancer Center; Research Cancer Center, Nantes University, Inserm U601, Nantes, France; Nuclear Medicine, Radiology, and Medical Oncology Units, Eugène Marquis Cancer Center; and Gynecologic Surgery Unit, University Hospital Pontchaillou, Rennes, France Address reprint requests to Caroline Rousseau, Service de Médecine Nucléaire, Centre René Gauducheau, Blvd Monod, 44805 Saint Herblain Cedex, France; e-mail: c-rousseau{at}nantes.fnclcc.fr
PURPOSE: This study aimed to assess prospectively the efficacy of sequential [18F]fluorodeoxyglucose positron emission tomography (FDG PET) to evaluate early response to neoadjuvant chemotherapy in stage II and III breast cancer patients. PATIENTS AND METHODS: Images were acquired with a PET/computed tomography scanner in 64 patients after administration of FDG (5 MBq/kg) at baseline and after the first, second, third, and sixth course of chemotherapy. Ultrasound and mammography were used to assess tumor size. Decrease in the standardized uptake value (SUV) with PET was compared with the pathologic response. RESULTS: Surgery was performed after six courses of chemotherapy and pathologic analysis revealed gross residual disease in 28 patients and minimal residual disease in 36 patients. Although SUV data did not vary much in nonresponders (based on pathology findings), they decreased markedly to background levels in 94% (34 of 36) of responders. When using 60% of SUV at baseline as the cutoff value, the sensitivity, specificity, and negative predictive value of FDG PET were 61%, 96%, and 68% after one course of chemotherapy, 89%, 95%, and 85% after two courses, and 88%, 73%, and 83% after three courses, respectively. The same parameters with ultrasound (US) and mammography were 64%, 43%, and 55%, and 31%, 56%, and 45%, respectively. Assessment of tumor response with US or mammography was never significant whatever the cutoff. CONCLUSION: Pathologic response to neoadjuvant chemotherapy in stage II and III breast cancer can be predicted accurately by FDG PET after two courses of chemotherapy.
Systemic therapy has been used for many years as primary treatment in stage II and III breast cancer.1,2 Early clinical response after two or three cycles of chemotherapy has been shown to be a predictor of pathologic complete remission and may therefore serve as a predictor for long-term outcome.3 The complete remission of all viable tumor tissue, as shown by a histopathologic examination of the removed breast and axillary tissue, is considered to indicate the complete eradication of locoregional disease by systemic medical treatment, and is widely proposed as a surrogate indication of the complete eradication of distant micrometastatic residual disease. This view is supported by the favorable long-term outcomes consistently seen in patients who achieve a pathologic complete remission after neoadjuvant systemic therapy.4,5 Chollet et al6 observed a histologically confirmed partial or complete regression in only 20% to 30% of clinical responders, suggesting that clinical and pathologic responses are not correlated. To minimize adverse effects of primary chemotherapy, nonresponders must be identified as early as possible. Current methods to assess response include physical examination, ultrasound (US), and mammography. Unfortunately, with these methods, no conclusion can be reached until three cycles of chemotherapy are completed.7 Because the change in tumor metabolism precedes the decrease in tumor size, positron emission tomography (PET) with [18F]fluorodeoxyglucose (FDG) should allow visualization of tumor response at an earlier stage than with conventional imaging methods.8,9 Two studies recently demonstrated in a limited number of patients that an early decrease in FDG uptake on serial FDG PET scans taken before and during chemotherapy can identify pathologic responders reliably.10,11 This study aimed to assess prospectively sequential FDG PET imaging as a means of early prediction of pathologic response to neoadjuvant chemotherapy in patients with stage II and III breast cancer.
Demography This prospective study was approved by the ethics committee and all patients gave their informed consent before enrollment. Sixty-four consecutive patients with newly diagnosed, noninflammatory, stage II and III breast cancer undergoing neoadjuvant chemotherapy were included. Diagnosis of invasive breast carcinoma was done by core needle biopsy in all patients and the stage of cancer was determined by x-ray, US, and bone scan investigations according to TNM classification (version 6).12 Exclusion criteria were the following: pregnancy, prior breast surgery, chemotherapy, or radiation therapy; known diabetes; age younger than 18 years; unwillingness or inability to undergo serial FDG PET scans; no tumor uptake at baseline; or ineligibility for surgery. Tumor size and location were established by US and mammography at baseline and after the first, third, and last course of chemotherapy. Mammography in two standard planes of imaging (mediolateral oblique and craniocaudal) was performed with the Mammomat Novation instrument (Siemens AG, Knoxville, TN). Both axillary and breast US were performed using an 11- to 13.5-MHz transducer (Sonoline Antares; Siemens AG, Erlanger, Germany). The neoadjuvant regimens administrated during this study, including doses and schedule, are summarized in Table 1. Surgery was performed less than 4 weeks after the last course of chemotherapy. Patients either underwent mastectomy (n = 14 of 64) or wide local excision (n = 50 of 64) with axillary lymph node dissection.
PET Imaging Tumors were imaged at baseline, after the first, second, and third course of chemotherapy, and before surgical excision of the primary tumor and axillary lymph nodes (Fig 1). Scans were performed with a Discovery LS PET/computed tomography (CT) scanner (General Electric Medical Systems, Milwaukee, WI). Patients had fasted 4 hours before PET acquisitions and blood glucose had to be less than 7 mmol/L before injection of 310 to 450 MBq (5 to 7 MBq/kg) of FDG. Intravenous injection in the opposite arm using a venous line (to prevent extravasation) was followed by a period of approximately 60 minutes, during which patients remained in a quiet room. No muscle relaxants were administered. Patients were allowed to breathe normally during PET and CT acquisitions. During acquisitions, patients were first in supine position with their arms raised above the head to pull their breasts away from the myocardium and facilitate examination of axillary regions. Then, whole-body image (from the proximal femur to the neck) with both arms along the body was acquired at baseline so that metastases would not be overlooked. Before the initial PET/CT scan, each patient's position (in three planes) was recorded (using a laser guide) to allow accurate repositioning for each subsequent scan.
PET data were acquired in two-dimensional mode for two bed positions (thoracic region) and for 5 minutes per bed position. PET images were reconstructed using CT data, for attenuation correction, using the ordered-subsets expectation maximization algorithm and without CT-based attenuation correction. Interpretation of PET data was done by two nuclear physicians blinded to clinical, radiologic, and pathologic findings and axillary lymph node status. A region of interest (ROI) of 5 to 10 mm was placed manually over the area of maximal activity on slices with the clearest definition of the tumor mass and in the adjacent slices. The standardized uptake value (SUV) was calculated based on the measured activity, decay-corrected injected dose, and patient body weight. This semiquantitative analysis was carried out for corresponding slices acquired before, during, and after treatment.
Clinical Assessment
Pathologic Assessment
Statistical Analysis
Patients and Pathologic Response Patients and tumor characteristics are listed in Table 2. The median primary tumor size was 40 mm (range, 10 to 100 mm). All patients with T1 tumors and only patients with T2N1 have cytologic confirmed ipsilateral axillary metastasis. Patients received four to six courses of primary chemotherapy and the median duration of treatments was 100 days (range, 78 to 127 days). Median tumor size after chemotherapy was 4 mm (range, 0 to 45 mm). After completion of chemotherapy, based on pathology findings, 36 patients were considered as responders. A Sataloff grade A response was identified in 10 patients, a grade B response was identified in 26 patients, and a grade C or D response was identified in 28 patients (Fig 2). A higher number of estrogen- and progesterone-positive patients were observed among responders (grade A + B) than nonresponders (grade C + D; P < .01 and P < .02). No difference was observed between responders and nonresponders regarding the HER2/neu status.
Treatment Response Assessed by Conventional Methods and Histopathology Using an arbitrary cutoff value of 60% of tumor size at baseline, the sensitivity, specificity, positive predictive value, and negative predictive value (NPV) after six courses of chemotherapy were 100%, 31%, 67%, and 100% with physical examination, 64%, 43%, 53%, and 55% with US, and 31%, 56%, 42%, and 45% with mammography, respectively. Conventional modalities (US and mammography) did not predict histologic response after three courses of chemotherapy (insufficient tumor size decrease; Fig 3).
SUV Before Neoadjuvant Chemotherapy Maximum and average uptake values within an ROI were found to be highly correlated (r = 0.92 at baseline, r = 0.87 after the first course, r = 0.80 after the second, r = 0.85 after the third, and r = 0.91 after the sixth course of chemotherapy; P < .00001). Therefore, we used the maximum SUV (SUVmax) normalized to blood glucose at the time of tracer injection for the following statistical analyses. The agreement between both readers was good, as shown by  = 0.86 (P < .0000001). The median SUVmax in patients who achieved a grade A response (16.6; range, 3.4 to 23.7) was significantly higher than that of patients with a grade B (8.4; range, 1.6 to 20.6; P = .045) and grade C + D response (5.3; range, 2.5 to 18.8; P = .003). Based on histology findings, SUVmax in primary lesions was higher (8.4) in ductal carcinomas than lobular carcinomas (3.7; P < .002).
Assessment of Treatment Response by PET PET studies performed after the first, second, third, and sixth course of chemotherapy were compared with the PET study at baseline, which was considered as the reference value. Figure 4 summarizes uptake values expressed as SUVmax. A significant difference was observed between FDG tumor uptake in the grade A, B, and C + D response groups (P < .0001). Interestingly, this difference was noticeable as early as after the first course of chemotherapy (P < .00001) and persisted until after the second and third course of chemotherapy (same degree of significance at both of these time points). After the last course of chemotherapy, the difference between tumor FDG uptake in these three groups was always observed but with a slightly lower degree of significance (P < .001). The grade A response group was characterized by a 60% decrease in SUVmax after the first course (Table 3). After the last course of chemotherapy, the SUVmax decreased sharply to background levels (ie, in surrounding normal breast tissue). Conversely, the grade C + D response group was characterized by a small decrease in SUVmax after the first course (16.3% of baseline SUVmax), followed by a small additional decrease in the SUVmax after the second and third courses of chemotherapy. After the last course, the SUVmax was always half of that at baseline. The grade B response group was characterized by intermediate results. After the first course, the decrease was lower than that observed in the grade A response group (35.6% of SUVmax at baseline). The SUVmax was then shown to decrease regularly after the second and third courses until the last course, when the SUVmax reached background levels.
We then sought to determine the threshold for optimal differentiation between responders and nonresponders. The receiver operating characteristic curve showed that the highest NPV was obtained after two cycles of chemotherapy, with a 40% decrease in the SUVmax measures at baseline (Table 4). With this cutoff value, 20 of 23 responders and 27 of 28 nonresponders were identified after the first course of therapy. The specificity was 96%, and pathologic response was predicted with an accuracy of 77%. After the second course of chemotherapy, the SUVmax decreased below the threshold value (40% of SUVmax at baseline) in 31 of 35 pathologic responders and 23 of 27 nonresponders, resulting in a sensitivity of 89%, specificity of 95%, and an NPV of 85%. Accuracy reached 87%.
If we compared NPV of conventional modalities with those of PET after two, three, and six courses of chemotherapy as shown in Figure 5, it was observed that NPV of PET was higher than for the other modalities since an early stage in the chemotherapy monitoring.
The higher rate of breast-conserving surgery is the most important advantage of preoperative chemotherapy in stage II and III breast cancer.1 In addition, it has been shown that after completion of chemotherapy, patients with minimal residual disease confirmed by pathology have a better prognosis than those with gross residual disease.4 Clearly, there is a need for correct preoperative identification of both responders and nonresponders. This prospective study demonstrated that patients with microscopic and macroscopic residual disease can be distinguished noninvasively by FDG PET as early as after the first course of chemotherapy, and even more effectively after the second course, with a sensitivity, specificity, and accuracy of 61%, 96%, 77%, and 89%, 95%, 87%, respectively. Although in nonresponders glucose metabolism reduced by approximately 50%, in responders SUV decreased to background levels.
FDG PET has been used previously to assess treatment response. Nine studies of treatment response monitoring have been published so far: eight of them studied the possibility of using PET to predict early response to neoadjuvant chemotherapy,8,10,11,16-21 whereas the last study22 examined the predictive value of PET after completion of chemotherapy. The number of FDG PET scans varied between three and four including baseline,11,16,20-22 compared with five in our study; all of these studies included a limited number of patients ( Smith et al20 found a higher baseline FDG uptake in patients with pathologic complete response than in patients who responded less. Similarly, a significant difference between baseline SUVmax in grade A, grade B, and grade C + D response groups was observed in our study. To assess FDG PET efficacy quantitatively, an SUV cutoff point decrease with regard to baseline must be selected. With a cutoff of 55% decrease, Schelling et al21 found an accuracy for response of 88% and 91% after the first and second course of chemotherapy, respectively. In our study, the cutoff SUV was selected based on the NPV of FDG uptake because medical oncologists require a high degree of certainty in the early identification of nonresponders due to the adverse effects of neoadjuvant chemotherapy. Consequently, we selected to wait for the second cycle of chemotherapy for prediction of response, based on a 40% decrease cutoff and an NPV of 85%, compared with 68% after the first course. The type of tumor is believed to limit pretreatment visualization of the primary tumor (eg, lobular carcinoma), which is also a problem with all other imaging modalities.23,24 Weaker correlation has been reported with microvessel density—a surrogate of angiogenesis25—and tumor cell density.26 In this study, most tumors were invasive ductal carcinomas; however, lobular carcinomas were identified on pretreatment PET images. The method used to define the ROI is of crucial importance in the monitoring of tumor FDG uptake during therapy. No consensus has been reached on the optimal type of ROI to be used to monitor response during therapy.27 Reproducibility, user independence, and to a lesser extent, simplicity, are important criteria to consider when choosing a method to define the ROI. In this study, the reproducibility of our method (with two observers) was demonstrated and the choice of the maximum pixel value within an ROI is a reasonable alternative to partial volume correction. Current methods to monitor tumor response to neoadjuvant chemotherapy (ie, physical examination and conventional imaging modalities) have limitations. Prior studies reported that nearly 50% of patients with a clinical complete response present with macroscopic residual disease at surgery, whereas 20% of patients with clinical partial response do not exhibit macroscopic tumor at surgery.28,29 Sensitivity and specificity of 81% and 38%, respectively, were reported for prediction of response to neoadjuvant chemotherapy by physical examination.30 We confirmed these results with a sensitivity and specificity of 100% and 31%, respectively. Mammography does not provide cross-sectional images, but rather projection images, which make evaluation of tumor size uncertain.31 US, as an adjunct to mammography, also has major limitations for the evaluation of tumor response, because it cannot differentiate a residual tumor from fibrosis.6 Results of our study confirmed all of these limitations, showing a poor NPV of 55% and 45% for US and mammography, respectively, after the last course of chemotherapy. All of these methods, however, do not allow differentiation between viable tumor tissue and fibrotic scar tissue.16 In summary, in contrast to mammography and US, FDG PET, using a simple imaging protocol, is able to provide early information on tumor response after the second course of neoadjuvant chemotherapy in stage II and III breast cancer. Early information about tumor response is extremely helpful in deciding the most appropriate therapeutic strategy. A prospective randomized study is needed that uses PET assessment after two cycles to determine a change to a different neoadjuvant regimen or to discontinue chemotherapy versus continuing chemotherapy in those patients without clinical or radiologic progression.
The authors indicated no potential conflicts of interest.
We thank the patients who participated in the study, and the nuclear medicine technologists at the René Gauducheau and Eugène Marquis Cancer Centers, N. Fleury, O. Zekri, E. Cerato, and M. Vinson.
published online ahead of print at www.jco.orgon November 6, 2006. Supported by Conseil Régional des Pays de Loire. Presented in poster format at the 41st Annual Meeting of the American Society of Clinical Oncology, May 13-17, 2005, Orlando, FL; and at the 52nd Annual Meeting of the Society of Nuclear Medicine, June 17-23, 2005, Toronto, Canada. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Fisher B, Brown A, Mamounas E, et al: Effect of preoperative chemotherapy on local-regional disease in women with operable breast cancer: Findings from National Adjuvant Breast and Bowel Project B-18. J Clin Oncol 15:2483-2493, 1997 2. Eltahir A, Heys SD, Huicheon AW, et al: Treatment of large and locally advanced breast cancers using neoadjuvant chemotherapy. Am J Surg 175:127-132, 1998[CrossRef][Medline] 3. Von Minckwitz G, Raab G, Caputo A, et al: Doxorubicin with cyclophosphamide followed by docetaxel every 21 days compared with doxorubicin and docetaxel every 14 days as preoperative treatment in operable breast cancer: The GEOARDUO study of the German Breast Group. J Clin Oncol 23:2676-2685, 2005 4. Bear HD, Anderson S, Smith RE, et al: Sequential preoperative or postoperative docetaxel added to preoperative doxorubicin plus cyclophosphamide for operable breast cancer: National Surgical Adjuvant Breast and Bowel Project Protocol B-27. J Clin Oncol 24:2019-2027, 2006 5. Wolmark N, Wang J, Mamounas E, et al: Preoperative chemotherapy in patients with operable breast cancer: Nine-year results from National Surgical Adjuvant Breast and Bowel Project B-18. J Natl Cancer Inst Monogr 30:96-102, 2001 6. Chollet P, Charrier S, Brain E, et al: Clinical and pathological response to primary chemotherapy in operable breast cancer. Eur J Cancer 3:862-866, 1997 7. Baum RP, Przetak Ch: Evaluation of therapy response in breast and ovarian cancer patients by positron emission tomography (PET). Q J Nucl Med 45:257-268, 2001[Medline] 8. 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 9. Avril N, Dose J, Jaenicke F, et al: Metabolic characterization of breast tumors with positron emission tomography using F-18 fluorodeoxyglucose. J Clin Oncol 14:1848-1857, 1996 10. Mankoff DA, Dunnwald LK, Gralow JR, et al: Changes in blood flow and metabolism in locally advanced breast cancer treated with neoadjuvant chemotherapy. J Nucl Med 44:1806-1814, 2003 11. Tiling R, Linke R, Untch M, et al: 18F-FDG PET and 99mTc-sestamibi scintimammography for monitoring breast cancer response to neoadjuvant chemotherapy: A comparative study. Eur J Nucl Med 28:711-720, 2001[CrossRef][Medline] 12. Sobin LH, Wittekind Ch: TNM Classification of Malignant Tumors (ed 6). New York, NY, Wiley-Liss, 2002 13. Ferriere JP, Assier I, Cure H, et al: Primary chemotherapy in breast cancer: Correlation between tumor response and patient outcome. Am J Clin Oncol 21:117-120, 1998[CrossRef][Medline] 14. Miller AB, Hoogstraten B, Staquet M, et al: Reporting results of cancer treatment. Cancer 47:207-214, 1981[CrossRef][Medline] 15. Sataloff DM, Mason BA, Prestipino AJ, et al: Pathologic response to induction chemotherapy in locally advanced carcinoma of the breast: A determinant of outcome. J Am Coll Surg 180:297-306, 1995[Medline] 16. 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 17. Bruce DM, Evans NT, Heys SD, et al: Positron emission tomography: 2-Deoxy-2-(18F)-fluoro-D-glucose uptake in locally advanced breast cancers. Eur J Surg Oncol 21:280-283, 1993[CrossRef] 18. Jansson T, Westlin JE, Ahlström H, et al: Positron emission tomography studies in patients with locally advanced and/or metastatic breast cancer: A method for early therapy evaluation? J Clin Oncol 13:1470-1477, 1995[Abstract] 19. Bassa P, Kim EE, Inoue T, et al: Evaluation of preoperative chemotherapy using PET with fluorine-18-fluorodeoxyglucose in breast cancer. J Nucl Med 37:931-938, 1996 20. Smith IC, Welch AE, Hutcheon AW, et al: Positron emission tomography using [18F]-fluorodeoxy-D-glucose to predict the pathologic response of breast cancer to primary chemotherapy. J Clin Oncol 18:1676-1688, 2000 21. Schelling M, Avril N, Nährig J, et al: Positron emission tomography using [18F]fluorodeoxyglucose for monitoring primary chemotherapy in breast cancer. J Clin Oncol 18:1689-1695, 2000 22. Kim SJ, Kim SK, Lee ES, et al: Predictive value of [18F]FDG PET for pathological response of breast cancer to neo-adjuvant chemotherapy. Ann Oncol 15:1352-1357, 2004 23. 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 24. Eubank WB, Mankoff DA: Evolving role of positron emission tomography in breast cancer imaging. Semin Nucl Med 35:84-99, 2005[CrossRef][Medline] 25. Bos R, van Der Hoeven JJ, van der Wall, et al: Biologic correlates of [18F]-fluorodeoxyglucose uptake in human breast cancer measured by positron emission tomography. J Clin Oncol 20:379-387, 2002 26. Avril N, Menzel M, Dose J, et al: Glucose metabolism of breast cancer assessed by 18F-FDG-PET: Histologic and immunohistochemical tissue analysis. J Nucl Med 42:9-16, 2001 27. Krak NC, Hoekstra OS, Lammertsma AA: Measuring response to chemotherapy in locally advanced breast cancer: Methodological considerations. Eur J Nucl Med 31:S103-S111, 2004[CrossRef] 28. Feldman LD, Hortobagyi GN, Buzdar AU, et al: Pathological assessment of response to induction chemotherapy in breast cancer. Cancer Res 46:2578-2581, 2000 29. Maini CL, Tofani A, Sciuto R, et al: Technetium-99mTc-MIBI scintigraphy in the assessment of neoadjuvant chemotherapy in breast carcinoma. J Nucl Med 38:1546-1551, 1997 30. Vranjesevic D, Filmont JE, Meta J, et al: Whole-body (18)F-FDG PET and conventional imaging for predicting outcome in previously treated breast cancer patients. J Nucl Med 43:325-329, 2002 31. Vinnicombe SJ, MacVicar AD, Guy RL, et al: Primary breast cancer: Mammographic changes after neoadjuvant chemotherapy, with pathologic correlation. Radiology 198:333-340, 1996 Submitted January 18, 2006; accepted September 21, 2006. This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2006 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
30) compared with our study. Based on baseline PET only (with 18F-labeled FDG and oxygen-15–labeled water), Mankoff et al
