|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2000 American Society for Clinical Oncology Positron Emission Tomography Using [18F]-Fluorodeoxy-D-Glucose to Predict the Pathologic Response of Breast Cancer to Primary ChemotherapyByFrom the John Mallard Scottish Positron Emission Tomography Center; Departments of Bio-Medical PhysicsSurgery, and Radiology, University of Aberdeen; and Departments of Oncology and Pathology, Grampian University Hospitals National Health Service Trust, Aberdeen, Scotland, United Kingdom. Address reprint requests to Ian Smith, MB, ChB, Department of Academic Radiology, University of Aberdeen, Foresterhill House Annex, Foresterhill, Aberdeen, AB25 2ZD Scotland, United Kingdom; email i.c.smith{at}abdn.ac.uk
PURPOSE: To determine whether [18F]-fluorodeoxy-D-glucose ([18F]-FDG) positron emission tomography (PET) can predict the pathologic response of primary and metastatic breast cancer to chemotherapy. PATIENTS AND METHODS: Thirty patients with noninflammatory, large (> 3 cm), or locally advanced breast cancers received eight doses of primary chemotherapy. Dynamic PET imaging was performed immediately before the first, second, and fifth doses and after the last dose of treatment. Primary tumors and involved axillary lymph nodes were identified, and the [18F]-FDG uptake values were calculated (expressed as semiquantitative dose uptake ratio [DUR] and influx constant [K]). Pathologic response was determined after chemotherapy by evaluation of surgical resection specimens. RESULTS: Thirty-one primary breast lesions were identified. The mean pretreatment DUR values of the eight lesions that achieved a complete microscopic pathologic response were significantly (P = .037) higher than those from less responsive lesions. The mean reduction in DUR after the first pulse of chemotherapy was significantly greater in lesions that achieved a partial (P = .013), complete macroscopic (P = .003), or complete microscopic (P = .001) pathologic response. PET after a single pulse of chemotherapy was able to predict complete pathologic response with a sensitivity of 90% and a specificity of 74%. Eleven patients had pathologic evidence of lymph node metastases. Mean pretreatment DUR values in the metastatic lesions that responded did not differ significantly from those that failed to respond (P = .076). However, mean pretreatment K values were significantly higher in ultimately responsive cancers (P = .037). The mean change in DUR and K after the first pulse of chemotherapy was significantly greater in responding lesions (DUR, P = .038; K, P = .012). CONCLUSION: [18F]-FDG PET imaging of primary and metastatic breast cancer after a single pulse of chemotherapy may be of value in the prediction of pathologic treatment response.
FIFTEEN PERCENT TO 25% of breast cancers are large (> 3 cm) or locally advanced (T3, T4, or TXN2) at the time of initial presentation.1,2 Patients with such tumors pose major therapeutic problems for surgeons and oncologists. First, these individuals manifest a high incidence of locoregional recurrence after surgery, and second, they have a poor prognosis, with many eventually succumbing to the effects of distant metastases.3,4 In order that the primary tumor may be downstaged before surgery and occult distant metastases abolished, primary chemotherapy (also known as neoadjuvant or induction chemotherapy) is now increasingly being used in the management of patients with large and locally advanced breast cancers.5,6 However, the survival of patients who require such treatment remains poor, because a significant proportion of these patients possess tumors that fail to respond to primary chemotherapy regimens.7,8 Studies have demonstrated that patients with unresponsive tumors may achieve an improved survival with the use of alternative and/or more prolonged courses of chemotherapy.9,10 Conventional methods of assessing tumor response rely on the documentation of morphologic changes in the neoplastic mass. Unfortunately, such changes are only apparent after the administration of several doses of toxic and expensive chemotherapy.11 Therefore, patients who would benefit from alternative treatments cannot be identified until a late point in their chemotherapy regimen. It would be of obvious benefit if patients with unresponsive tumors could be identified at a much earlier stage, thereby avoiding the use of ineffective treatment. Several studies have demonstrated the ability of functional imaging techniques to detect subclinical alterations in tumor physiology and biochemistry resulting from efficacious therapy.12-21 These alterations may occur long before a morphologic change in the tumor mass is apparent. Positron emission tomography (PET) using the labeled glucose analog [18F]-fluorodeoxy-D-glucose ([18F]-FDG) is a functional imaging technique that is thought to allow the quantification of in vivo cellular glycolysis.22-24 To date, only two studies have been published that have evaluated the ability of [18F]-FDG PET to assess the response of breast cancers to chemotherapy.25,26 Both of these studies demonstrated that PET could identify clinical response to treatment at a much earlier stage in the therapeutic regimen than was possible using conventional procedures. However, as far as we are aware, no study has been published that has specifically evaluated the ability of PET to predict pathologic tumor response. It is now apparent that the pathologic, rather than the clinical, response of breast cancers to primary chemotherapy is of considerable prognostic importance.8,27,28 Consequently, large-scale primary chemotherapy studies are currently underway with the aim of improving pathologic breast cancer response rates by the administration of prolonged treatment regimens.8,29 It is thought that, to accurately quantify cellular glycolysis using PET, stringent experimental conditions must be satisfied. These include a knowledge of the [18F]-FDG activity within the arterial blood pool, estimation of the volume of distribution to which the radiopharmaceutical has been administered, correction for errors arising as a consequence of the finite spatial resolution of the imaging modality, and determination of the rate at which [18F]-FDG accumulates within the area being studied. In practice, this necessitates the use of time-consuming dynamic imaging protocols, the collection of sequential arterial blood samples, and the application of complex methods of PET data analysis. The use of such laborious techniques may prevent the widespread clinical application of fully quantitative PET imaging. Furthermore, the translation of PET-derived values of [18F]-FDG uptake to the cellular glycolytic rate requires that broad assumptions be made regarding [18F]-FDG kinetics and metabolism.23,30,31 In our opinion, insufficient evidence exists to support the validity of such assumptions when applied to breast cancer. Therefore, in the present study, we have deliberately avoided the temptation of implying that the rate of [18F]-FDG uptake approximates that of cellular glycolysis. Rather, we have attempted to show that changes in [18F]-FDG uptake can provide a marker of response to therapy. In view of these considerations, the main aim of this study was to determine whether [18F]-FDG PET using a simple, noninvasive imaging protocol that may be easily applied in the busy clinical setting could be used to predict pathologic breast cancer response to primary chemotherapy. In addition, we hypothesized that the response of metastatic breast cancer within axillary lymph nodes may be of even greater prognostic importance than the response of the tumor that remains localized in the breast. The secondary aim of this study, therefore, was to assess the ability of [18F]-FDG PET to predict the pathologic response of locally metastatic breast cancer within axillary lymph nodes to primary chemotherapy.
Patients and Treatment Patients with newly diagnosed, noninflammatory, large (> 3 cm), or locally advanced (T3, T4, or TXN2) breast cancers who had agreed to participate in an ongoing study comparing the efficacy of an anthracycline-based primary chemotherapy regimen (cyclophosphamide, doxorubicin, vincristine, and prednisolone [CVAP]) with that of primary docetaxel (Doc) were eligible for inclusion in this investigation. All patients initially received four doses of CVAP chemotherapy (cyclophosphamide 1,000 mg/m2 intravenously [IV], doxorubicin 50 mg/m2 IV, and vincristine 1.5 mg/m2 IV followed by oral prednisolone 40 mg for 5 days, every 21 days). After this initial treatment, the clinical response of the primary tumor was assessed and graded according to standard International Union Against Cancer criteria.32 If there had been progression or stasis of disease, patients received four further doses of Doc (100 mg/m2 IV followed by oral prednisolone 100 mg for 5 days, every 21 days). However, if there had been a partial or complete response to treatment thus far, patients were randomized to receive either four further pulses of CVAP or four further pulses of Doc (as above). After completion of chemotherapy, all patients received appropriate surgery to remove the primary breast tumor and sample the axillary lymph nodes. The Joint Ethical Committee of the University of Aberdeen and Grampian Health Board approved the study protocols. All patients were required to provide informed consent before study entry.
Evaluation of Tumor Response All surgical resection specimens obtained after completion of chemotherapy were evaluated independently by two consultant pathologists (I.D.M. and S.P.) in order to determine the degree of pathologic tumor response of the primary breast lesion, the presence of axillary lymph node metastases, and their pathologic responses to treatment. The pathologists were unaware of details regarding clinical tumor response and the results of PET data analyses. Pathologic tumor response was graded according to previously described criteria.33,34 For primary breast lesions, a partial pathologic response (pPR) was deemed to have occurred when macro- and microscopically evident residual neoplastic tissue demonstrated features consistent with chemotherapy-induced damage. A macroscopic complete pathologic response (pCR-macro) was defined as the absence of macroscopically visible tumor. A microscopic complete pathologic response (pCR-micro) was defined as the histologic absence of invasive tumor cells. Tumors that failed to satisfy the criteria for pPR or pCR were classified as exhibiting a pathologic nonresponse. Lymph nodes within axillary dissection specimens were deemed to have exhibited pCR-micro if all tissue examined was free of neoplastic cells and the lymph node architecture demonstrated histologic evidence of previous tumor involvement (Figs 1 and 2.)
PET Imaging Protocol Tumors were imaged at the John Mallard Scottish PET Centre, Aberdeen, Scotland, immediately before the first, second, and fifth doses of chemotherapy and before surgical excision of the primary tumor and axillary lymph nodes. A CTI/Siemens (Knoxville, TN) ECAT EXACT 31 scanner was used. It has an axial field of view of 10.8 cm and produces 31 tomographic planes simultaneously with a spatial resolution of 6 to 8 mm in the axial and both transaxial directions. To standardize blood glucose levels, patients were required to fast for at least 6 hours before PET imaging (many fasted overnight). Before the commencement of the initial PET scan, each patient’s position (in three planes) within the imager was recorded (using the laser guides within the imager referenced to bony landmarks), thus allowing accurate positioning for each subsequent scan. Imaging was performed over a 20-cm span below the midpoint of the ipsilateral clavicle to the breast lesion. This required data collection at two bed positions. Ten-minute transmission data sets were acquired at each bed position before injection of approximately 185 MBq of [18F]-FDG into a dorsal foot vein on the contralateral side to the breast lesion. This relatively low dose of activity was administered because patients were to receive four PET scans during the period of their treatment regimen. The [18F]-FDG was synthesized on site using a method developed by Haaparanta et al.35 After injection of the radiopharmaceutical agent, dynamic emission data were acquired over the axilla for 60 minutes. The dynamic imaging protocol consisted of eight 15-second time frames, four 30-second time frames, one 60-second time frame, one 300-second time frame, and five 600-second time frames. This was followed by a 10-minute static acquisition of emission data over the site of the breast lesion. A venous blood sample was drawn before the administration of [18F]-FDG, for determination of the plasma glucose level. PET images were acquired in a two-dimensional mode and reconstructed using the standard filtered backprojection algorithm with a Hanning window and a cutoff at the Nyquist frequency. The reconstruction array consisted of 31 contiguous transaxial slices, each of 128 x 128 voxels. The voxels were cubic, with sides measuring 3.375 mm. Corrections were applied for randoms, scatter, and dead time.
PET Image Analysis
The dynamic data were analyzed using the Patlak method.36 The arterial input function was obtained using an operator-selected region of interest (ROI) over a major arterial blood pool (the aortic arch), as validated by Germano et al.37 To correct for any partial volume effects caused by the size of the arterial vessel, the whole time activity curve was scaled to match the activity in a venous blood sample taken 30 minutes after injection. The influx constant (K) was calculated for each voxel of the image, which resulted in a parametric map of [18F]-FDG uptake.38 The dose uptake ratio (DUR) was also calculated for each voxel from the final frame of the dynamic data, using the following equation:
PET data for each patient were viewed as a set of 62 contiguous transaxial slices (31 dynamically derived parametric maps and 31 static images). Pretreatment PET images of the breasts (static images) and axillae (parametric images of [18F]-FDG uptake and static images derived from the final frame of the dynamic data set) were visually inspected to determine the presence and extent of focal accumulations of [18F]-FDG consistent with the presence of neoplastic lesions (Figs 3 and 4). ROIs were manually drawn around each lesion, and the maximum pixel value of DUR or K within the ROI was recorded. For each patient, subsequently obtained PET data were analyzed by applying the ROI identified in the pretreatment image to precisely the same physical volume.
Statistics Data were analyzed using SPSS for Windows version 9.0 (SPSS Inc, Chicago, IL). Alpha was set at 0.05 (two-tailed). The means of the primary and metastatic tumor DUR and K values were calculated for each PET data set. Distributions were examined to determine whether they approximated normality. Mean DUR and K values were compared using Student’s t test. Least squares linear regression was used to compare DUR values with pathologic tumor grade. Nominal variables were compared using the  2 test. The relationships between clinical and pathologic response, tumor size and [18F]-FDG uptake (DUR or K), and DUR and K values were evaluated by means of Pearson’s correlation coefficient (rxy). The ability of PET to predict the ultimate pathologic response of primary breast cancers (using the percentage change in DUR after the first dose of chemotherapy) was analyzed using receiver operating characteristic (ROC) curves.39 ROC curves were generated and analyzed using the ROCKIT program (IBM-compatible version 0.9B, March 1998, Department of Radiology, University of Chicago, Chicago, IL). P values less than .05 were considered significant.
Patient Characteristics and Tumor Response Thirty patients were enrolled onto the study. One patient had two separate assessable primary breast lesions. Therefore, data regarding a total of 31 primary breast cancers were available for analysis. Details of clinical disease stage at diagnosis, treatment administered, and tumor responses are listed in Table 1. The mean age of this study population was 49 years (range, 31 to 72 years). The overall clinical (cPR and cCR) and pathologic (pPR and pCR) response rates after completion of the chemotherapy regimen were 90% and 58%, respectively. Thirty-eight percent of the breast lesions had undergone a pCR-macro and 26% of lesions had undergone a pCR-micro. cCR was an accurate indication of complete macroscopic ( 2 = 7.89, P < .01) and microscopic (2 = 10.11, P < .01) pathologic response. Higher tumor grade was not associated with a greater incidence of pCR-macro or pCR-micro. Mean patient weight and blood glucose levels did not differ significantly at each episode of PET imaging. All patients received eight doses of primary chemotherapy over a mean regimen duration of 157 days (range, 135 to 184 days). The dose-intensity of each chemotherapy regimen administered did not differ significantly and was not related to the ultimate clinical or pathologic tumor response.
PET Imaging A total of 98 PET scans were performed on the 30 patients enrolled onto the study. Due to relocation of the PET imaging facility, two patients did not receive a PET scan before their second pulse of chemotherapy, 11 patients failed to receive a scan before their fifth pulse of treatment, and nine did not receive a scan after their last pulse of treatment. None of the patients refused PET imaging. Visual inspection of the pretreatment PET images revealed the primary breast tumors in all patients. Focal abnormalities within the axilla were clearly identified in all 11 patients who were subsequently shown to have pathologic evidence of axillary lymph node involvement (Figs 3 and 4). The [18F]-FDG uptake values of the primary and locally metastatic lesions derived by PET image analysis before the first, second, and fifth doses and after the last dose of chemotherapy are listed in Table 2.
Pretreatment DUR and K Data Primary breast lesions that demonstrated an overall pathologic response (pPR and pCR) after chemotherapy had a mean pretreatment DUR of 0.091 (range, 0.030 to 0.193). This value did not differ significantly (t = 0.14, P = .887) from that for lesions that failed to demonstrate a pathologic response (mean DUR, 0.088; range 0.032 to 0.217). The mean pretreatment DUR value obtained from the primary breast lesions that failed to undergo a pCR-macro to chemotherapy was 0.082 (range, 0.032 to 0.217) (Table 3). This value did not differ significantly (t = 1.25, P = .222) from that obtained from the breast lesions that did achieve a pCR-macro to chemotherapy (mean DUR, 0.106; range, 0.030 to 0.193). However, the mean pretreatment DUR value obtained from the eight primary breast lesions that achieved a pCR-micro to chemotherapy (mean DUR, 0.123; range, 0.031 to 0.193) was significantly higher (t = 2.19, P = .037) than that obtained from the 23 lesions that failed to undergo a similar response (mean DUR, 0.079; range, 0.03 to 0.217). When the study population was divided into two subgroups according to the chemotherapy regimen administered, no significant difference in the mean pretreatment DUR values was found between pathologically responsive and nonresponsive breast lesions (Table 4). The mean pretreatment DUR value in the primary breast lesions that achieved a clinical response after four or eight pulses of chemotherapy did not differ significantly from the mean value measured in the lesions that failed to undergo a clinical response.
Regression analysis revealed that there was a significant positive linear correlation between pretreatment DUR and the histologic grade of the primary tumor (DUR, r = .42, P = .019) (Fig 5).
Changes in DUR Compared With Primary Breast Cancer Response The primary breast lesions that were deemed to have undergone a clinical response (cCR or cPR) after completion of the chemotherapy regimen demonstrated a significantly greater mean reduction in DUR (t = 2.55, P = .017) after a single dose of chemotherapy than those lesions that failed to achieve such a response (Table 3). Primary breast cancers that responded pathologically to treatment (pPR or pCR) demonstrated a significantly (t = 2.66, P = .013) greater reduction in DUR after the first dose of chemotherapy than those cancers that failed to exhibit any evidence of pathologic response. For primary tumors that ultimately achieved a pCR-macro, a significant reduction in the mean DUR was observed after the first (t = 3.24, P = .003), fourth (t = 2.12, P = .014), and eighth (t = 2.32, P = .031) pulses of chemotherapy when compared with the DUR measured in less responsive tumors. The eight primary breast cancers that achieved a pCR-micro after completion of the treatment regimen also demonstrated a significant reduction in their mean DUR values after the first (t = 3.55, P = .012), fourth (t = 3.76, P = .001), and final pulses (t = 3.07, P = .006) of chemotherapy. Similar results were observed irrespective of the treatment regimen administered (Table 4). Expressed as a percentage of the pretreatment value, breast cancers that had an overall pathologic response (pPR or pCR) demonstrated a mean reduction in DUR of 27.8%. Primary cancers that underwent a pCR-macro and pCR-micro demonstrated a mean reduction in DUR of 38.4% and 43.7%, respectively, after a single pulse of chemotherapy. These reductions were all significantly greater than those observed for less responsive tumors (for overall pathologic response, t = 3.56, P = .001; for pCR-macro, t = 3.76, P = .001; for pCR-micro, t = 3.50, P = .002). Using a 10% reduction in DUR as a cutoff, PET imaging after a single pulse of chemotherapy predicted the overall pathologic response (pPR, pCR-macro, or pCR-micro) of the primary breast lesion with a sensitivity of 82.4% and specificity of 66.7% (Fig 6). When a 20% reduction in DUR after a single dose of chemotherapy was applied, it was possible to predict a complete pathologic response (pCR-macro and pCR-micro) with a sensitivity of 90% and specificity of 74%. ROC curve analyses demonstrated that the area under the curve for the prediction of overall pathologic response was 0.856 (SE, ± 0.07), the area under the curve for the prediction of pCR-macro was 0.891 (SE, ± 0.07), and the area under the curve for the prediction of pCR-micro was 0.929 (SE, ± 0.06). The differences between the areas under the curves for the prediction of complete pathologic response (microscopic and macroscopic) and that for the prediction of overall pathologic response were not significant (Fig 7).
Changes in DUR and K Compared With Metastatic Tumor Response Eleven patients had pathologic evidence of axillary lymph node involvement after completion of chemotherapy. The metastatic tumors in three of these patients showed a pCR-micro to treatment. The mean reductions in DUR after the first pulse of chemotherapy in responsive and nonresponsive tumors were 44% and 24.3% respectively (Table 5). The mean reductions in K after the first pulse of chemotherapy in responsive and nonresponsive tumors were 55.3 and 20%, respectively. Therefore, the mean reductions in both DUR and K were significantly greater in responding metastatic tumors compared with nonresponding tumors (for DUR, t = 2.48, P = .038; for K, t = 3.26, P = .012). DUR and K values were significantly correlated before treatment (rxy = .97, P < .001) and after the first (rxy = .98, P < .001), fourth (rxy = .089, P = .02), and last (rxy = .99, P < .001) doses of chemotherapy.
Primary Lesion Size and Measured [18F]-FDG Uptake Changes in the clinical size of the primary breast cancers were determined by comparing the sizes of the lesions (the greatest assessable tumor diameter) before treatment with their sizes after the first, fourth, and last doses of chemotherapy. Pathologic tumor size was determined by measuring the extent of the residual tumor visible macroscopically. Before treatment, the mean tumor size was 4.9 cm (range, 1 to 8 cm). The mean tumor size after the fourth dose of chemotherapy was 2.3 cm (range, 0 to 7cm), and after the last dose of chemotherapy, it was 1.2 cm (range, 0 to 8 cm). No appreciable change in the clinical size of the primary breast lesions was apparent after the first dose of chemotherapy. The mean pathologic size of residual tumor was 1.6 cm (range, 0 to 6 cm). The DUR values obtained from primary lesions after completion of chemotherapy were significantly correlated with clinical (rxy = .443, P = .045) and pathologic (rxy = .650, P = .005) tumor size. However, DUR values obtained before the start of treatment and after the first and fourth doses of chemotherapy were not significantly correlated with clinical tumor size.
It is now increasingly apparent that PET can be used to visualize primary breast cancers and tumor-involved axillary lymph nodes with a high degree of accuracy.40 It is not surprising, therefore, that all 31 large or locally advanced breast cancers studied were identified on the pretreatment PET images. In addition, the pretreatment PET images of the 11 patients who subsequently were proven to have pathologic evidence of axillary lymph node metastases demonstrated corresponding focal areas of [18F]-FDG uptake. The principle aims of this study were to determine whether [18F]-FDG PET could be used to quantify changes in rate of [18F]-FDG accumulation by tumors as a result of the use of cytotoxic drugs and to determine whether the magnitude of such changes could predict pathologic breast cancer response to primary chemotherapy. Several studies have demonstrated the ability of PET to document clinical cancer response to antineoplastic therapies.12,14,16,41,42 However, none of these studies has actually specifically evaluated the ability of PET to predict the eventual pathologic tumor response. As described above, we have not implied that changes in [18F]-FDG uptake by neoplastic lesions approximate changes in the glycolytic rate. Although we acknowledge that under certain exact circumstances it may be possible to study cellular glycolysis using [18F]-FDG PET, the approach adopted in the present study was to determine whether a relatively simple method of assessing [18F]-FDG uptake (not glycolysis) by breast cancers could be used to predict response to chemotherapy. Primary chemotherapy is widely used in the management of patients with large and locally advanced breast cancers.43,44 The complete pathologic resolution of the tumor after chemotherapy is known be of considerable prognostic importance27,43,45,46 and frequently does not correlate with observed clinical response.8,27,28,47 Therefore, this study has specifically addressed the question of whether PET can predict the pathologic rather than clinical response of breast cancers to primary chemotherapy. The presence of regional lymph node metastases is an indication of occult systemic tumor spread and therefore is the most important prognostic factor in patients with breast cancer.48,49 It has been suggested that tumor cells that metastasize have a lower sensitivity to systemic chemotherapy than the malignant cells that remain in situ within the breast.50-52 Therefore, the response of metastatic tumor within regional lymph nodes probably reflects the response of occult disseminated disease and may be of greater prognostic significance than the response of the primary tumor.28,53-55 PET is one of the most reliable methods of noninvasively demonstrating the presence of axillary lymph node metastases40,56; therefore, we also assessed the ability of this imaging modality to predict the response within tumor-involved axillary lymph nodes. Several studies have now demonstrated that the visualization of focal accumulations of [18F]-FDG within small neoplastic lesions,38,57 such as tumor-involved lymph nodes, is best achieved by creating parametric maps of [18F]-FDG enhancement. This approach proved similarly successful in the present study. It may be hypothesized that tumors with higher glycolytic rates, and therefore higher rates of [18F]-FDG uptake, may achieve a superior response to antineoplastic therapy.58,59 If this is the case, then a single PET scan before initiation of chemotherapy may be sufficient to predict cancer response. The present study has demonstrated a significant difference in the mean pretreatment rates of [18F]-FDG uptake in primary breast lesions that achieved a pCR-micro when compared with less responsive cancers and in responding and nonresponding metastatic tumors. We acknowledge that these observations are derived from small numbers of patients and therefore their clinical significance is uncertain; however, we do believe this area merits further study. Linear regression analysis did reveal a significant positive correlation between tumor grade and the pretreatment rate of [18F]-FDG uptake, a finding consistent with those from other investigators.60-63 However, in this study, pathologic response was not associated with a higher tumor grade. Primary breast cancers that achieved a pPR or pCR (macroscopic and microscopic) demonstrated a significantly greater reduction in the rate of [18F]-FDG uptake after a single pulse of chemotherapy than those cancers that failed to achieve a pathologic response (Table 3). This observation was consistent irrespective of the chemotherapy regimen administered (Table 4). Therefore, these results suggest that [18F]-FDG PET may be useful in predicting a pathologic cancer response to a variety of chemotherapy agents at an early stage in a treatment regimen. When a 20% reduction in [18F]-FDG uptake was used to identify those cancers that would ultimately achieve a pCR, a correct prediction was achieved with a sensitivity of 90% and a specificity of 74% after a single dose of chemotherapy. A principle therapeutic aim of primary chemotherapy is to achieve a pCR within the breast lesion. It is well known that malignant cells may possess or develop a resistance to the cytotoxic effects of the initial chemotherapy agents used.64,65 Patients with such cancers often benefit from the administration of alternative drugs.66 The results of the present study suggest that PET may be useful in identifying such patients at an early stage in the treatment regimen. Focal areas of elevated [18F]-FDG uptake, within the axillae, were visible on the PET images obtained from 11 patients before and after the first pulse of chemotherapy. All 11 patients eventually were proven to have pathologic evidence of axillary lymph node metastases. The architecture of the lymph nodes obtained by surgical dissection of the axillae in three of these patients demonstrated features consistent with that of previous tumor involvement. Therefore, these patients were deemed to have experienced a pCR of locally metastatic tumor. Analysis of PET data obtained from the presumed pathologically responsive axillary lymph node metastases showed the rates of [18F]-FDG uptake (DUR) to have fallen by 40%, 43%, and 50% after the first pulse of treatment, compared with a corresponding mean percentage fall in [18F]-FDG uptake (DUR) of 24% (range, 7% to 41%) demonstrated in ultimately nonresponsive nodes. The rates of [18F]-FDG uptake described by DUR and K values were significantly correlated. Furthermore, all lymph node metastases were evident by visual inspection of both the pretreatment parametric maps of [18F]-FDG uptake and the static images derived from the final frame of the dynamic acquisition sequence. Although the numbers of patients are small, these results do suggest that it may be possible to use semiquantitative PET assessment of [18F]-FDG uptake to predict the pathologic response of metastatic tumor within axillary lymph nodes after a single dose of primary chemotherapy. These observations, together with those pertaining to the primary tumor, may be clinically relevant if PET is to be used as a routine method of predicting cancer response to therapy, because semiquantitative assessment of [18F]-FDG uptake is technically much simpler and involves much less imaging time than a full quantitative assessment. Before treatment, clinical examination suggested that nine patients had axillary lymph node metastases; however, pathologic evaluation of axillary dissection specimens in three of these patients failed to reveal evidence of tumor involvement. Although it is possible that tumor deposits within the axillae of these patients were eradicated by primary chemotherapy, the lymph nodes examined pathologically did not reveal architectural changes consistent with previous tumor involvement. Furthermore, clinical examination of axillae is known to be an unreliable method of assessing of pathologic lymph node status67 and inferior to assessment using [18F]-FDG PET.68 The authors of several recent reports28,53,69 have suggested that axillary lymph node status may change during the course of a primary chemotherapy regimen and that knowledge of this change may be of considerable prognostic significance. The results of this study indicate that PET may be used before and during a chemotherapy regimen to assess the status of locoregional lymph nodes during treatment. However, it will not be possible to assess the pertinence and prognostic significance of this potentially extremely valuable information until a period of time has passed that is sufficient to allow survival analyses to be conducted. Although we have not attempted to quantify the glycolytic rate of the cancers we have studied, several methodologic issues must still be considered when interpreting the data presented in this study. Quantitative analyses using tomographic imaging techniques are known to be susceptible to errors arising from partial volume effects.30,70 These errors occur when the size of the lesion being assessed is smaller than the resolution of the imaging modality. When such a situation arises, the lesion may be visualized as a discrete focal accumulation of activity, but the measured activity within the lesion will be somewhat less than the actual activity. Therefore, unless such partial volume effects are appropriately corrected, quantitative analyses obtained from a breast lesion less than 1.5 cm in size may be considered inaccurate.31,71 The significant correlations observed in the present study between final clinical and pathologic tumor size and final DUR give a clear indication that the measured [18F]-FDG accumulation within the residual tumor at this time point is being altered by partial volume effects. However, with the exception of one cancer (lesion 2 of patient no. 30), all primary breast lesions studied were initially greater than 1.5 cm in size. Furthermore, clinical examination before the second PET scan demonstrated that no alterations in tumor size had occurred during the first treatment cycle. Therefore, we believe it is unlikely that partial volume effects would have significantly altered the predictive value of the early PET results pertaining to primary breast lesions. Reductions in measured [18F]-FDG activity for lesions within axillary lymph nodes and those derived from the primary breast lesions at later time points during the chemotherapy regimen will be a partial manifestation of the degree of morphologic tumor response. It is known that elevated blood glucose levels reduce cellular uptake of [18F]-FDG.72-74 Therefore, care must be taken to ensure that [18F]-FDG uptake values are normalized to blood glucose levels, particularly if members of the study population are hyperglycemic at the time of imaging. We believe that it is unlikely the results of this study have been influenced by variations in plasma glucose concentration, because all patients included were nondiabetic and adequately fasted (confirmed by blood glucose estimation) at the time of PET imaging. Furthermore, we found no significant difference in the mean plasma glucose levels of any patient immediately before PET imaging. In conclusion, therefore, the results of this study indicate that [18F]-FDG PET, using a simple imaging protocol, may be of considerable value in the early prediction of pathologic response of both primary and locally metastatic breast cancer to chemotherapy. Further studies, involving larger patient numbers, are currently underway in an attempt to validate this hypothesis.
Supported by the Cancer Research Campaign, London, United Kingdom. We gratefully acknowledge the effort and assistance provided to us while conducting this study by the staff of the Behavioural Oncology Unit within the University of Aberdeen.
1. Cancer Research Campaign: Breast cancer: UK. Cancer Research Campaign Fact Sheet 6.1, London, United Kingdom, Cancer Research Campaign, 1996 2. Carter CL, Allen C, Henson DE: Relation of tumor size, lymph node status, and survival in 24,740 breast cancer cases. Cancer 63:181-187, 1989[Medline] 3. Haagensen CD, Stout AP: Carcinoma of the breast: I. Results of treatment. Ann Surg 116:801-815, 1942[Medline] 4. Haagensen CD, Stout AP: Carcinoma of the Breast: II. Criteria of operability. Ann Surg 118:859-870, 1032-1051, 1943 5. Bonadonna G, Valagussa P, Zucali R, et al: Primary chemotherapy in surgically resectable breast cancer. CA Cancer J Clin 45:227-243, 1995[Abstract] 6. Wang H, Lo S: Future prospects of neoadjuvant chemotherapy in treatment of primary breast cancer. Semin Surg Oncol 12:59-66, 1996[Medline] 7. Clavel M, Catimel G: Breast cancer: Chemotherapy in the treatment of advanced disease. Eur J Cancer 29A:598-604, 1993
8.
Fisher B, Brown A, Mamounas E, et al: Effect of preoperative chemotherapy on loco-regional disease in women with operable breast cancer: Findings from National Surgical Adjuvant Breast and Bowel Project B-18. J Clin Oncol 15:2483-2493, 1997 9. Heys SD, Eremin JM, Sarkar TK, et al: Role of multimodality therapy in the management of locally advanced carcinoma of the breast. J Am Coll Surg 179:493-504, 1994[Medline] 10. Frei E III, Canellos GP: Dose, a critical factor in cancer chemotherapy. Am J Med 69:585-594, 1980[Medline] 11. Moskovic EC, Mansi JL, King DM, et al: Mammography in the assessment of response to medical treatment of large primary breast cancer. Clin Radiol 47:339-344, 1993[Medline] 12. Price P, Jones T: Can positron emission tomography (PET) be used to detect subclinical response to cancer therapy? The EC PET Oncology Concerted Action and the EORTC PET Study Group. Eur J Cancer 31A:1924-1927, 1995 13. Chaiken L, Rege S, Hoh C, et al: Positron emission tomography with fluorodeoxyglucose to evaluate tumor response and control after radiation therapy. Int J Radiat Oncol Biol Phys 27:455-464, 1993[Medline]
14.
Findlay M, Young H, Cunningham D, et al: Noninvasive monitoring of tumor metabolism using fluorodeoxyglucose and positron emission tomography in colorectal cancer liver metastases: Correlation with tumor response to fluorouracil. J Clin Oncol 14:700-708, 1996 15. Nagata Y, Yamamoto K, Hiraoka M, et al: Monitoring liver tumor therapy with 18[F]FDG positron emission tomography. J Comput Assist Tomogr 14:370-374, 1990[Medline]
16.
Reisser C, Haberkorn U, Dimitrakopoulou-Strauss A, et al: Chemotherapeutic management of head and neck malignancies with positron emission tomography. Arch Otolaryngol 121:272-276, 1995 17. Minn H, Kangas L, Knuutila V, et al: Determination of 2-fluoro-2-deoxy-D-glucose uptake and ATP level for evaluating drug effects in neoplastic cells. Res Exp Med 191:27-35, 1991[Medline] 18. Jansson T, Westlin JE, Ahlstrom 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. Rieber A, Zeitler H, Rosenthal H, et al: MRI of breast cancer: Influence of chemotherapy on sensitivity. Br J Radiol 70:452-458, 1997[Abstract] 20. Buscombe JR, Cwikla JB, Thakrar DS, et al: Scintigraphic imaging of breast cancer: A review. Nucl Med Commun 18:698-709, 1997[Medline] 21. Lagalla R, Caruso G, Finazzo M: Monitoring treatment response with color and power doppler. Eur J Radiol 27:s149-s156, 1999 22. Phelps ME, Huang SC, Hoffman EJ, et al: Tomographic measurement of local cerebral glucose metabolic rate in humans with (F-18)2-fluoro-2-deoxy-D-glucose: Validation of method. Ann Neurol 6:371-388, 1979[Medline]
23.
Hawkins RA, Choi Y, Huang SC, et al: Quantitating tumour glucose metabolism with FDG and PET. J Nucl Med 33:339-344, 1992 (editorial)
24.
Strauss LG, Conti PS: The applications of PET in clinical oncology. J Nucl Med 32:623-664, 1991
25.
Bassa P, Kim E, Inoue T, et al: Evaluation of preoperative chemotherapy using PET with fluorine-18-fluorodeoxyglucose in breast cancer. J Nucl Med 37:931-938, 1996
26.
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
27.
Feldman LD, Hortobagyi GN, Buzdar AU, et al: Pathological assessment of response to induction chemotherapy in breast cancer. Cancer Res 46:2578-2581, 1986 28. Fisher B, Bryant J, Wolmark N, et al: Effect of preoperative chemotherapy on the outcome of women with operable breast cancer. J Clin Oncol 16:2672-2685, 1998[Abstract] 29. Smith IC, Walker LG, Hutcheon AW, et al: Chemotherapy in breast cancer: Enhanced response with docetaxel. Eur J Surg Oncol 24:616, 1998 (abstr)
30.
Keyes JW: SUV: Standard uptake or silly useless value? J Nucl Med 36:1836-1839, 1995
31.
Avril N, Bense S, Ziegler SI, et al: Breast imaging with fluorine-18-FDG PET: Quantitative image analysis. J Nucl Med 38:1186-1191, 1997 32. Miller AB, Hoogstraten B, Staquet M, et al: Reporting results of cancer treatment. Cancer 47:207-214, 1981[Medline] 33. Sharkey FE, Addington SL, Fowler LJ, et al: Effects of preoperative chemotherapy on the morphology of resectable breast carcinoma. Mod Pathol 9:893-900, 1996[Medline] 34. Aktepe F, Kapucuoglu N, Pak I: The effects of chemotherapy on breast cancer tissue in locally advanced breast cancer. Histopathology 29:63-67, 1996[Medline] 35. Haaparanta M, Bergman J, Solin O, et al: Remote controlled system for the routine synthesis of 18F-2-fluoro-L-2 deoxy-D-glucose. J Nucl Med 21:823, 1985 (suppl) (abstr) 36. Patlak CS, Blasberg RG, Fenstermacher JD: Graphical evaluation of blood-to-brain transfer constants from multiple-time uptake data. J Cereb Blood Flow Metab 3:1-7, 1983[Medline]
37.
Germano G, Chen BC, Huang SC, et al: Use of the abdominal aorta for arterial input function determination in hepatic and renal PET. J Nucl Med 33:613-620, 1992 38. Messa C, Choi Y, Hoh CK, et al: Quantification of glucose utilization in liver metastases: Parametric imaging of FDG uptake with PET. J Comput Assist Tomogr 16:684-689, 1992[Medline]
39.
Travis LB, Curtis RE, Glimelius B, et al: Bladder and kidney cancer following cyclophosphamide therapy for Non-Hodgkin’s lymphoma. J Natl Cancer Inst 87:524-530, 1995 40. Smith IC, Welch AE, Chilcott F, et al: Gamma emission imaging in the management of breast disorders. Eur J Surg Oncol 24:320-329, 1998[Medline]
41.
Ichiya Y, Kuwabara Y, Otsuka M, et al: Assessment of response to cancer therapy using fluorine-18-fluorodeoxyglucose and positron emission tomography. J Nucl Med 32:1655-1660, 1991 42. Huovinen R, Leskinen-Kallio S, Nagren K, et al: Carbon-11-methionine and PET in evaluation of treatment response of breast cancer. Br J Cancer 67:787-791, 1993[Medline] 43. Ellis PA, Smith IE: Primary chemotherapy for early breast cancer. Cancer Treat Rev 22:437-450, 1996[Medline]
44.
Bonadonna G, Valagussa P, Brambilla C, et al: Primary chemotherapy in operable breast cancer: Eight-year experience at the Mian Cancer Institute. J Clin Oncol 16:93-100, 1998 45. Eltahir A, Heys SD, Hutcheon AW, et al: Treatment of large and locally advanced breast cancers using neo-adjuvant chemotherapy: Recurrence and survival. Am J Surg 175:127-132, 1998[Medline]
46.
Esteva FJ, Hortobagyi G: Integration of systemic chemotherapy in the management of primary breast cancer. The Oncologist 3:300-313, 1998 47. Gazet JC, Coombes RC, Ford HT, et al: Assessment of the effect of pretreatment with neoadjuvant therapy on primary breast cancer. Br J Cancer 73:758-762, 1996[Medline]
48.
McCready DR, Hortobagyi GN, Kau SW, et al: The prognostic significance of lymph node metastases after preoperative chemotherapy for locally advanced breast cancer. Arch Surg 124:21-25, 1989 49. Fisher B, Bauer M, Wickerham L, et al: Relation of the number of positive axillary nodes to the prognosis of patients with primary breast cancer. Cancer 52:1551-1557, 1983[Medline] 50. Rubens RD: The management of locally advanced breast cancer. Br J Cancer 65:145-147, 1992[Medline] 51. Fidler IJ, Hart IR: The origin of metastatic heterogeneity in tumours. Eur J Cancer 17:487-494, 1981 52. Skipper HE: Kinetics of mammary tumor cell growth and implications for therapy. Cancer 28:1479-1499, 1971[Medline] 53. Kuerer HM, Sahin A, Hunt KK, et al: Incidence and impact of documented eradication of breast cancer axillary lymph node metastases before surgery in patients treated with neoadjuvant chemotherapy. Ann Surg 230:72-78, 1999[Medline] 54. Kuerer HM, Newman LA, Buzdar AU, et al: Residual metastatic axillary lymph nodes following neoadjuvant chemotherapy predict disease-free survival in patients with locally advanced breast cancer. Am J Surg 176:502-509, 1998[Medline] 55. Botti C, Vici P, Lopez M, et al: Prognostic value of lymph node metastases after neoadjuvant chemotherapy for large-sized operable carcinoma of the breast. J Am Coll Surg 181:202-208, 1995[Medline] 56. Murray AD: Imaging of the axilla in breast cancer. Imaging 9:239-246, 1997
57.
Zasadny KR, Wahl RL: Enhanced FDG-PET tumor imaging with correlation-coefficient filtered influx-constant images. J Nucl Med 37:371-374, 1996
58.
Higashi K, Clavo AC, Wahl RL: Does FDG uptake measure proliferative activity of human cancer cells? In vitro comparison with DNA flow cytometry and titrated thymidine uptake. J Nucl Med 34:414-419, 1993 59. Oza AM, Tannock IF: Breast cancer: Clinical relevance of breast cancer biology. Hematol Oncol Clin North Am 8:1-14, 1994[Medline] 60. Alavi JB, Alavi A, Chawluk J, et al: Positron emission tomography in patients with glioma: A predictor of prognosis. Cancer 62:1074-1078, 1988[Medline] 61. Adler LP, Blair HF, Williams RP, et al: Grading liposarcomas with PET using [18F]FDG. J Comput Assist. Tomogr 14:960-962, 1990 62. Di Chiro G: Positron emission tomography using [F18]fluorodeoxyglucose in brain tumors. Invest Radiol 22:360-371, 1987[Medline] 63. Crippa F, Seregni E, Agresti R, et al: Association between [18F]fluorodeoxyglucose uptake and postoperative histopathology, hormone receptor status, thymidine labelling index and p53 in primary breast cancer: A preliminary observation. Eur J Nucl Med 25:1429-1434, 1998[Medline]
64.
Gottesman MM: How cancer cells evade chemotherapy: Sixteenth Richard and Hinda Rosenthal Foundation award lecture. Cancer Res 53:747-754, 1993
65.
Goldstein LJ, Galski H, Fojo A, et al: Expression of a multidrug resistance gene in human cancers. J Natl Cancer Inst 81:116-124, 1989
66.
Hortobagyi GN: Treatment of breast cancer. N Engl J Med 339:974-984, 1998 67. Wallace IWJ, Champion HR: Axillary nodes in breast cancer. Lancet 1:217-218, 1972[Medline] 68. Smith IC, Ogston KN, Whitford P, et al: Staging the axilla in breast cancer: Accurate in vivo assessment using positron emission tomography with 2-(fluorine-18)-fluoro-2-deoxy-D-glucose. Ann Surg 228:220-227, 1998[Medline] 69. Mamounas EP: Overview of National Surgical Adjuvant Breast Project neoadjuvant chemotherapy studies. Semin Oncol 25:31-35, 1999 70. Hoffman EJ, Huang S, Phelps ME: Quantitation in positron emission computed tomography: 1. Effect of object size. J Comput Assist Tomogr 3:299-308, 1979[Medline]
71.
Avril N, Dose J, Janicke F, et al: Metabolic characterization of breast tumours with positron emission tomography using F-18 fluorodeoxyglucose. J Clin Oncol 14:1848-1857, 1996
72.
Langen KJ, Braun U, Kops ER, et al: The influence of plasma glucose levels on fluorine-18-fluorodeoxyglucose uptake in bronchial carcinomas. J Nucl Med 34:355-359, 1993
73.
Wahl RL, Henry CA, Ethier SP: Serum glucose: Effects on tumor and normal tissue accumulation of 2-[F-18]-fluoro-2-deoxy-D-glucose in rodents with mammary carcinoma. Radiology 183:643-647, 1992
74.
Diederichs CG, Staib L, Glatting G, et al: FDG PET: Elevated plasma glucose reduces both uptake and detection rate of pancreatic malignancies. J Nucl Med 39:1030-1033, 1998 Submitted January 21, 1999; accepted December 20, 1999.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2000 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
