- Email: [email protected]
Measuring the “Unmeasurable”
Measuring the ‘‘Unmeasurable:’’ Assessment of Bone Marrow Response to Therapy Using FDG-PET in Patients with Lymphoma Behnaz Goudarzi, MD, Heather A. Jacene, MD, Richard L. Wahl, MD Rationale and Objectives: To determine if anatomically ‘‘nonmeasurable’’ disease in bone marrow (BM) is assessable for response to therapy by [18F]-2-fluoro-2-deoxy-D-glucose (FDG) positron emission tomography (PET)/computed tomography (CT). Materials and Methods: FDG PET/CT images of 27 patients with lymphoma, FDG-avid bone marrow (BM) lesions, and $1 FDG-avid, tumor-involved lymph node (LN) at baseline were retrospectively reviewed. FDG uptake in target LNs and BM foci was determined preand posttherapy using the standardized uptake value corrected for lean body mass (SULmean). Size of the same target LNs was measured pre- and posttherapy on CT. Percentage decreases of LN size and LN and BM SUL were calculated. Response was classified according to revised International Workshop Criteria (IWC) with and without modification for metabolic evaluation of BM and correlated to overall survival. Statistical analyses were performed using paired t-tests, Pearson correlation coefficients, and z-tests. Results: LN size, LN SULmean, and BM SULmean were significantly higher pre- versus posttherapy (2337 mm2 1810 vs. 309 mm2 323; 6.94 4.96 vs. 1.02 1.00; and 6.81 4.58 to 1.84 1.58, all P < .001, respectively). After therapy, significant correlation was found between percentage declines of LN size and SULmean of LNs (r = 0.84, P < .001) or BM (r = 0.56, P = .002) and SULmean of LN and BM (r = 0.76, P < .001). Including a metabolic assessment of BM correctly altered overall response assessment in 5/27 (19%) patients and better predicted overall survival than revised IWC. Conclusion: Anatomically ‘‘unmeasurable’’ BM infiltration with lymphoma behaves similarly to LN disease after therapy and is ‘‘measurable’’ by FDG PET/CT. FDG PET/CT is valuable for monitoring tumor response in ‘‘measurable’’ disease and BM, which was previously considered ‘‘unmeasurable’’ by anatomical imaging. Key Words: FDG; lymphoma; PET/CT; bone marrow; response assessment. ªAUR, 2010
T
umor involvement of bone marrow (BM) in patients with lymphoma defines a higher stage and portends a worse prognosis. Biopsy is the current standard for diagnosing lymphomatous infiltration of BM. Sensitivity for identifying disease in BM increases 10%–20% by multiple sampling; however, even with negative bilateral biopsies, patients may subsequently prove to have BM disease (1,2). Bone and BM lesions are considered ‘‘nonmeasurable’’ in therapy response criteria, which are primarily based on anatomic imaging. The terms ‘‘measurable’’ and ‘‘nonmeasurable’’ were introduced in the Response Evaluation Criteria in Solid Tumors (RECIST) approach (3). Per RECIST, ‘‘measurable lesions’’ can be accurately measured in at least one dimension with the longest diameter being $10 or 20 mm depending on computed tomography (CT) technique.
Acad Radiol 2010; 17:1175–1185 From the Division of Nuclear Medicine, Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University, 601 North Caroline Street, JHOC 3223, Baltimore, MD 21287-0817 (B.G., H.A.J., R.L.W.). Received February 18, 2010; accepted May 1, 2010. Supported in part by a grant from NIH/NCI (3 P30 CA006973-43S2 Abeloff/Wahl for IRAT: Imaging Response Assessment Teams in Cancer Center Supplement). Address correspondence to: R.L.W. e-mail: [email protected] ªAUR, 2010 doi:10.1016/j.acra.2010.05.001
Bone lesions and several others (eg, ascites, pleural/pericardial effusions, leptomeningeal disease, lymphangitis cutis/pulmonis, cystic lesions) are specifically considered as ‘‘truly nonmeasurable’’ per RECIST. Although no direct measurements of tumor burden are taken, response of ‘‘nonmeasurable’’ lesions is ‘‘considered’’ when assigning a posttreatment response classification. Numerous studies have demonstrated that positron emission tomography (PET) with [18F]-2-fluoro-2-deoxy-Dglucose (FDG) detects BM tumor involvement in patients with lymphoma, often in the presence of normal CT scans (4,5), and provides complementary information to BM biopsy (6–17). The discordancy rate between FDG PET and BM biopsy is ~20% (6,11–13,15–17). False-negative PET scans for BM disease occur with lower tumor burden (<35%) in BM, low-level FDG uptake in the primary nodal disease, or specific histologic subtype (12,17). Positive PET scans in BM with negative BM biopsy (ie, false-positive PET scans) have also been reported. This may be due to sampling error, and disease is often detected with PETdirected rebiopsy of focal FDG uptake distant from the site of the initial negative BM biopsy (6,11,16). In early studies, diffuse FDG uptake throughout the BM was another cause of false-positive PET scans. At least in Hodgkin’s lymphoma (HL), it is now recognized that diffuse, 1175
GOUDARZI ET AL
TABLE 1. Patient Characteristics (Total Group, n = 27) Characteristics Sex (M/F) 20/7 Age (mean SD) 45 21 years Histologies HL 9 Nodular sclerosing 7 Mixed cellularity 1 Not otherwise specified 1 NHL 18 Large B cell 16* Follicular 2 Pretherapy PET indication Initial staging 18 Restaging 9 Interval therapy 21 Chemotherapy aloney Radioimmunotherapy 4 Donor lymphocyte infusion 1 Stem cell transplant 1 Timing of posttherapy PET Mid-therapy After 2 cycles of chemotherapy 4 After 3 cycles of chemotherapy 10 13 After completion of therapyz Days between (mean SD) Pre- and posttherapy PET 95 50 Last therapy day and posttherapy PET 31 27 Last therapy day and pretherapy, restaging PET 100 54 HL, Hodgkin’s lymphoma; NHL, non-Hodgkin’s lymphoma; PET, positron emission tomography. *Including 2 transformed from low-grade follicular NHL. y Chemotherapy regimens consisted of R-CHOP in 12 and R-ICE in 1 with diffuse large B-cell NHL. The patient with Burkitt’s NHL received CHOP. HL regimens included ABVD (n = 2); ABVE-PC (n = 2); BCVD, gemcitabine/cisplatin in 1 each and 1 unknown regimen. z n = 7 after chemotherapy; n = 4 after radioimmunotherapy; n = 1 each after donor lymphocyte infusion and stem cell transplant.
homogeneous BM uptake at initial staging may be due to tumor infiltration but is more likely the result of reactive changes (6,18). The widely accepted International Workshop Criteria (IWC) (19) for assessing response of non-Hodgkin’s lymphoma (NHL) to treatment were revised to include an assessment of changes in tumor metabolism using FDG PET after treatment (IWC+PET) (20). In IWC+PET, biopsy remains the standard test for detecting lymphoma in BM before and after therapy (20). This is despite recommendations that clearly increased (multi)-focal FDG uptake in BM be interpreted as positive for lymphoma in companion guidelines for interpreting PET scans (21). One potential limitation of PET for evaluation of the BM in the posttherapy setting is that diffuse FDG uptake in reconstituting BM from endogenous or exogenous hematopoietic growth factors may obscure residual focal lesions (22–24). 1176
Academic Radiology, Vol 17, No 9, September 2010
Most previous investigations evaluated FDG uptake in BM lesions qualitatively and focused on disease detection and concordance with biopsy. This retrospective observational study was performed to compare the ability of FDG PET/ CT as a noninvasive imaging method to determine treatment response in anatomically ‘‘nonmeasurable’’ lesions (BM) versus measurable lesions (lymph nodes) by semiquantitative analysis and to determine the effect of including a metabolic evaluation of BM in the IWC+PET response assessment on overall survival. We hypothesized that BM involvement with tumor would be detectable, metabolically measurable with the standardized uptake value (SUV), would respond in a similar manner to anatomically ‘‘measurable’’ nodal disease, and would provide valuable prognostic information beyond that routinely available.
MATERIALS AND METHODS Our institutional review board approved this study under an expedited review and waived the requirement for informed consent. Patient Selection
PET/CT reports from 1100 patients with lymphoma imaged between June 2003 and April 2006 were retrospectively reviewed. Twenty-seven met the following inclusion criteria and were selected as the study population: diagnosis of HL or NHL; unifocal or multifocal FDG-avid lesions reported in BM (with or without cortical involvement); $1 involved, anatomically measurable lymph node (LN) or mass lesion detected on pretherapy PET/CT scan obtained; mid- or posttherapy PET/CT scan. Eighteen patients presented at initial diagnosis, whereas nine were studied at the time of recurrence or persistent disease because a restaging scan was required for possible change in management. Patient characteristics and interval therapies are shown in Table 1. PET/CT Imaging
Patients were fasting at least 4 hours and had serum glucose levels <200 mg/dL before intravenous injection of a weight-based dose of FDG (0.22 mCi/kg, 8.14 MBq/kg). Oral contrast (1.3% Readi-cat, EZEM, Lake Success, NY) was administered. After an approximately 60-minute tracer uptake phase, a combined PET/CT scan (Discovery LS or ST; GE Healthcare, Waukesha, WI) was obtained from mid-skull to midfemur level or beyond if necessary. Whole-body CT, without intravenous contrast, was performed first, with 4- or 16-slice multidetector helical scanners. CT parameters were specific for the particular scanner, but are summarized for the Discovery ST and LS: 120–140 kVp, weight-based amperage (20–250 mA), 0.5 and 0.8 seconds per CT rotation, pitch of 0.984:1 and 1.5:1, and 3.75 and 5 mm reconstructed slice
Academic Radiology, Vol 17, No 9, September 2010
thickness, respectively. Emission data were acquired in the two-dimensional mode for 5 minutes per bed position and were reconstructed using the ordered-subset expectation maximization algorithm (two iterations, 21 or 28 subsets), Gaussian post- filter with a 128 128 matrix, and CT attenuation correction. Posttherapy PET/CT scans were performed midchemotherapy, after two (n = 4) or three (n = 10) cycles in 14 patients, and after the completion of therapy in 13 patients (n = 7 chemotherapy; n = 4 radioimmunotherapy; n = 1 donor lymphocyte infusion; n = 1 myeloablative stem cell transplant). Times between pre- and posttherapy scans and last treatment and posttherapy scans are listed in Table 1. Image Interpretation
Two nuclear medicine physicians with experience in PET/ CT reviewed the original images for the 27 patients meeting study criteria, first independently and then in consensus. PET, CT, and fused PET/CT images were viewed on a single display screen using Xeleris software (GE Healthcare, Waukesha, WI). On the CT part of the pretherapy PET/CT images, up to six FDG-avid LNs were selected on the basis of their size (lesions with the longest diameters) and their suitability for accurate repeated measurements by imaging techniques. One hundred and thirty LNs were located in the following regions: neck (n = 16), axilla (n = 20), para-aortic/aortocaval (n = 13), mediastinum (n = 16), supraclavicular (n = 3), subpectoral (n = 3), hilum (n = 3), paravertebral (n = 2), pelvic (n = 13), inguinal (n = 7), upper abdomen (n = 13), and other (n = 21). The product of the long and short axis diameters for each LN was calculated and baseline sum of products of the diameters (SPD) determined. The same LNs were measured on the CT part of the posttherapy PET/CT scan and posttherapy SPD calculated. Large LN masses were included in the analysis if they had definitive borders. Liver, spleen, and lung masses were excluded. For the metabolic semiquantitative analysis, a 1-cm circular region of interest (ROI) was drawn around the most intense area of FDG accumulation in the same LNs or LN masses measured on CT. The mean and maximum (max) standardized uptake values corrected for lean body mass (SUL) within the 1-cm ROI were determined for each LN pre- and posttherapy. If a LN completely resolved on the posttherapy scan, background SUL was determined in the region of previous tumor. SULmean and SULmax were determined in up to six different foci of increased FDG uptake in BM. If the focus of FDG uptake completely resolved on the posttherapy scan, background SUL was determined in the region of the prior focus. Average Hounsfield units in a 1-cm diameter ROI were determined on CT corresponding to the location of the foci of increased FDG in BM lesions before and after therapy. Normal BM metabolic uptake was determined in the lumbar vertebrae by drawing a 1-cm diameter ROI.
ASSESSING MARROW RESPONSE WITH FDG-PET
It was not possible to histologically confirm all areas of increased FDG uptake as BM disease, but lesions were chosen for evaluation considering the overall pattern of disease on the baseline PET and correlative anatomic imaging. The results regarding lymphomatous infiltration of the BM on FDG PET/CT scanning were compared with biopsy and histology when available or other imaging modalities, such as magnetic resonance imaging (MRI), CT, or bone scan. Posttherapy Response Assessments
Posttherapy response was first classified according to IWC+PET (20). A complete response (CR) required a negative FDG PET scan and a negative BM biopsy (if positive pretherapy). For a partial response (PR), SPD of LNs must have declined by $50% and at least one LN remained PET positive posttherapy by qualitative visual assessment. In the setting of PET positivity in LNs posttherapy, BM biopsy results were ‘‘irrelevant’’ for response classification (ie, a PR would have been assigned if the BM biopsy was positive or negative) (20). Progressive disease (PD) was defined as $50% increase in SPD or size of one LN or any new FDG-avid lesion that was confirmed by histology or other modalities. New foci of FDG uptake in BM was not considered PD because BM biopsy is the standard procedure for assessment of BM and for all new previously uninvolved sites, histologic confirmation is necessary (19,20). IWC+PET response was then reclassified adding a qualitative visual assessment of metabolic response of BM lesions after therapy (PET+BM). In addition to a CR by IWC+PET, a CR by PET+BM required FDG uptake in BM lesions to be less than baseline and less than or equal to background BM. A CR or PR by IWC+PETwas considered a PR by PET+BM if there was residual focal FDG uptake in BM lesions that was less than baseline but greater than or equal to background BM. PD by PET+BM was defined as for IWC+PET but also as present if the intensity of an FDG-avid focus in BM visually increased or a new BM focus of FDG uptake consistent with a lymphomatous pattern was seen. Statistical Analyses
Statistical analyses were performed using S-Plus (Insight Corporation, Seattle, WA), Systat (Cranes Software, Karnataka, India), and/or Microsoft Excel (Microsoft, Seattle, WA). Differences of means were compared using two-tailed paired t-tests and correlations explored using Pearson correlation coefficients and z-tests. P values # .05 were considered statistically significant. Agreement between IWC+PET and PET+BM response assessments was evaluated with the kappa statistic (k). Overall survival (OS) was defined as time between pretherapy FDG PET/CT scan and death from any cause. OS was correlated with response using Kaplan-Meier curves and log-rank tests. 1177
GOUDARZI ET AL
Academic Radiology, Vol 17, No 9, September 2010
TABLE 2. Lymph Node Size and Lymph Node and Bone Marrow FDG Uptake Before Therapy After Therapy P Value Lymph node (n = 130) SPD (size, mm2) SULmean SULmax Bone marrow lymphoma (n = 91) SULmean SULmax Normal bone marrow SULmean SULmax
2337 1810 6.94 4.96 8.74 6.19
309 323 1.02 1.00 1.29 1.32
<.001 <.001 <.001
6.81 4.58 8.26 5.2
1.84 1.58 2.16 1.95
<.001 <.001
0.85 0.33 1.00 0.34
0.93 0.32 1.12 0.37
.28 .37
SPD, sum of products of the diameters; SULmean, mean standardized uptake value; SULmax, maximum standardized uptake value.
RESULTS Mean SULmean of LNs and LN size were significantly higher pre- versus posttherapy (6.94 4.96 vs. 1.02 1.00 and 2337 mm2 1810 vs. 309 mm2 323, P < .001, respectively; Table 2). SULmean of BM foci decreased significantly posttherapy (6.81 4.58 to 1.84 1.58, P < .001). An example is shown in Figures 1A and 1B. Similar results were obtained analyzing SULmax (Table 2). Normal BM metabolic activity in the vertebrae was unchanged with treatment (Table 2). No significant differences were found in HU of BM lesions before and after therapy (data not shown). Strong positive correlations were observed between percentage decreases of LN SPD and LN SULmean after therapy (r = 0.84, P < .001) and LN SULmean and BM SULmean (r = 0.76, P < .001) (Figs 2a,b). There was a positive correlation between percentage decrease of LN SPD and BM SULmean after therapy (r = 0.56, P = .002) (Fig 2c). Similar strong correlations were observed when evaluating percentage decreases of SULmax of the LN or BM and LN SPD (LN SPD vs. LN SULmax: r = 0.84, P < .0001; LN SULmax vs. BM SULmax: r = 0.79, P < .0001; LN SPD vs. BM SULmax: r = 0.59, P = .001). Comparing pre- versus posttherapy scans, no significant differences were found in several patient and technical parameters known to affect SUL, including blood glucose levels (95 13 vs. 99 13 mg/dL, P = .22), FDG uptake phase time (64 10 vs. 66 13 minutes, P = .45), and lean body mass (59 10 vs. 59 11 kg, P = .11). A summary of histologic and imaging correlation/confirmation is presented in Figure 3. Twenty-four patients had standard BM aspirations and biopsies performed at the iliac crest at a median of 2 days before the PET/CT scan (range, 156 days before to 63 days after the PET/CT scan). Four patients had positive biopsies for lymphomatous involvement in BM at the iliac crest. In 11 patients with a positive FDG PET/CT in BM but a negative iliac crest biopsy, contempo1178
raneous studies (range 14 days before PET/CT to 39 days after, median 0 days) revealed BM infiltration at the site of abnormal FDG uptake on PET as follows: one with PETdirected BM biopsy, one with biopsy before PET based on x-ray, one with bone scan and CT scans, three with bone scans, two with CT, and three with MRI. In the other nine patients, further imaging studies were suggestive of BM involvement in four and BM involvement was not confirmed in five. Directed BM biopsy was the only site of biopsy in 3 of 27 patients. In these cases, preceding imaging studies suggested bone/BM abnormalities and biopsies were performed before FDG PET/CT imaging. In all three cases, there was intense FDG uptake corresponding to abnormal CT findings on bone windows, greater than would be expected from biopsy alone. Overall, nine patients had a positive BM biopsy pretherapy. BM biopsy was repeated after therapy in six of these patients and all were negative. Nineteen of 27 (70%) patients achieved a CR after therapy, 5 (18%) had a PR, and 3 (12%) had PD by IWC+PET. IWC+PET and PET+BM response assessments (Table 3) agreed in 22 of 27 cases (81%, k = 0.67; Fig 4). In five patients (19%), response was classified as CR by IWC+PET but PR by PET+BM (Fig 5). Four of these five response assessments were made on mid-therapy posttreatment scans and one was based on posttransplant FDG PET/CT scan. For these five patients, three had iliac crest biopsies only before therapy. Median follow-up time was 30.2 months (range 4.1–77.8 months). OS was longer for patients who achieved a CR by IWC+PET and PET+BM (Fig 6). Of the five patients reclassified from CR to PR by PET+BM, two progressed at sites of residual FDG uptake in BM and died of progressive NHL 26.2 and 20.3 months after pretherapy PET. One achieved a CR after continuing with standard chemotherapy and was disease-free 19.1 months after pretherapy PET. Two had therapy intensified mid-treatment to stem cell transplant and converted to CR on posttherapy PET. One relapsed in lymph nodes 3 months posttransplant and died 12.3 months after the pretherapy scan. The other received additional posttransplant external beam radiation therapy to the BM lesion and was disease free 44.1 months after the pretherapy scan. Based on the available clinical history, 11 of 27 patients received hematopoietic growth factors a median of 12 days (range 3–43 days) before posttherapy PET. Of these 11, 5 were felt to visually have diffusely increased uptake in BM from the growth factors without any definite foci of more intense uptake to suggest residual BM disease. Eleven patients did not receive hematopoietic growth factors and receipt of growth factors is unknown in five patients, although one patient’s scan suggested recent receipt of growth factors. Three of the five patients with residual foci of uptake in BM posttherapy did not receive hematopoietic growth factors before posttherapy PET, one was unknown, and one had focal BM uptake above background suggesting residual disease in BM despite receiving growth factors 11 days before midtherapy PET scan (Fig 5).
Academic Radiology, Vol 17, No 9, September 2010
ASSESSING MARROW RESPONSE WITH FDG-PET
Figure 1. (a) Lymph node involvement before and after therapy is shown on these scans (computed tomography [CT], left; positron emission tomography [PET]/CT, middle; PET, right). Before therapy, the lymph node (arrows) measured 28 16 mm and had a mean standardized uptake value (SULmean) of 8.11. After therapy, the lymph node (crosshairs) measured 16 8 mm and had a SULmean of 0.72. (b) Bone marrow involvement in the sternum (arrowheads) is shown for the same patient (CT, left; PET/CT, middle; PET, right). Before therapy, SULmean in the BM lesion was 6.76 and after therapy it was 0.84.
1179
GOUDARZI ET AL
Academic Radiology, Vol 17, No 9, September 2010
Figure 2. Correlations between percent change of anatomic and metabolic parameters. There are strong positive correlations between (a) percent change in sum of products of the diameters (SPD) of LN and percent change in lymph node (LN) mean standardized uptake value (SULmean) after therapy (r = 0.84, P < .001), and (b) percent change in LN SULmean and bone marrow (BM) SULmean after therapy (r = 0.76, P < .001). (c) There is a positive correlation between the percent change in LN SPD and BM SULmean after therapy (r = 0.56, P = .002).
DISCUSSION Bone lesions have been considered ‘‘truly nonmeasurable’’ in criteria for monitoring response of tumor to therapy (RECIST and IWC). Similar to IWC, the RECIST guideline was recently revised (version 1.1); in RECIST 1.1, special consideration is given to bone lesions (25). The identifiable soft-tissue components of lytic bones lesions may be considered measurable if they meet all other criteria for measurable disease and can be evaluated by CT or MRI. PET scan is specifically considered an inadequate imaging technique to measure bone lesions. The results of our report suggest otherwise. For this investigation, we chose to evaluate the use of FDG PET for monitoring the response of BM disease in patients with lymphoma because this population relatively frequently has BM disease, has disease that responds to
Figure 3. Flow chart of histologic and imaging correlation/confirmation of FDG-avid bone marrow (BM) lesions. PET, positron emission tomography; CT, computed tomography; MRI, magnetic resonance imaging.
1180
therapy, and is now routinely assessed with FDG PET before and after therapy. Our discussion primarily focuses on PET for monitoring response of BM disease in patients with lymphoma, but the concepts should be considered for broader applicability to other tumor types and response criteria beyond the IWC for lymphoma. The recently proposed PET response criteria in solid tumors (PERCIST) 1.0 criteria allow for inclusion of BM disease in the treatment response assessment, just like other foci of FDGavid disease (26). In our study population, we successfully measured FDG uptake in regions of BM apparently infiltrated with lymphoma before and after therapy using the SUL parameter. The data presented support our hypotheses that changes in metabolic activity of lymphoma in the BM are in the same direction and magnitude as in nodal disease. To the best of our knowledge, no previous systematic studies have demonstrated this relationship offering a possible explanation for the lack of inclusion of assessment of BM disease in previous PET-based response criteria. The significance of semiquantitative uptake determinations in nodal lymphoma remains controversial. Even more limited data exist regarding the use of semiquantitative analyses of FDG uptake for foci of lymphomatous infiltration of BM. Schaefer et al reported a mean SUV of 6.26 3.22 in BM lesions (16) and our results are similar. Posttherapy SULmean of BM lesions decreased to 1.84 1.58 in our study. Schaefer et al did not compare pre- and posttherapy SUVs (16). Some authors have attempted to define ‘‘cutoff ’’ values for diagnosing residual disease or predicting prognosis in lymphoma (27,28). A sensitivity of 86% and specificity of 100% for the detection of residual/recurrent lymphoma by PET/CT was reported using a cutoff SUVmax of 2.5 (27). In 58 patients with HL or NHL who achieved a CR unconfirmed after therapy, FDG PET had a high accuracy for predicting early recurrence using a threshold SUVmax of 3 for the presence of recurrent lymphoma (28). Given the variability in cutoff values for level of FDG uptake, we did not attempt to define such cutoff values for LN or BM lesions in our patient population.
Academic Radiology, Vol 17, No 9, September 2010
TABLE 3. Responses After Therapy
IWC, International Workshop Criteria; PET, positron emission tomography; BM, bone marrow; CR, complete response; PR, partial response.
The IWC+PET response criteria rely on a visual qualitative assessment of changes in FDG uptake between baseline and posttherapy scans (20). Any additional benefit of semiquantitative analyses for characterizing response to therapy has yet to be definitively demonstrated, but preliminary data suggest this to be true (29). We categorized patients’ response to therapy according to IWC+PET criteria in this study, but we did not present these results versus changes in BM uptake because the majority of our patients had CR (70%) after therapy. In this study, each patient was used as his or her own control and we compared pre- and posttherapy SUL and also correlated percent changes rather than crude values. Unlike chemotherapy alone, hematopoietic growth factors cause a diffuse increase in FDG uptake in normal BM (22– 24,30–32), most markedly near the time of administration. The uptake declines, but can remain above baseline for 4 weeks after short-acting growth factors (22). A time course
ASSESSING MARROW RESPONSE WITH FDG-PET
of uptake after long-acting growth factors is not wellestablished. This is problematic because residual focal lesions in the BM may not only be obscured by the diffuse uptake but the differences in time between hematopoietic growth factor administration and PET can result in heterogeneity in FDG uptake in the BM between patients. In this study, diffuse FDG uptake in BM resulting from hematopoietic growth factors was noted in 6 of 27 patients posttheraphy, 5 with FDG uptake less than background BM in the locations of BM disease. Five achieved a CR in their nodal disease and, at last follow-up, median survival was 44 months. The sixth patient achieved a PR in nodal disease and died 13 months after the PET scan. We caution that application of a metabolic assessment of the BM (qualitative or quantitative) is limited in the setting of hematopoietic growth factors and may depend on the timing of the posttherapy PET scan after hematopoietic growth factor administration. However, we observed a ‘‘flip-flop’’ phenomenon (33), photopenia in the location of previous disease compared to stimulated normal BM in two cases (Fig 7), which was reassuring for the absence of metabolically active disease in BM. Primary malignant bone tumors and many other benign bone lesions have been reported to accumulate FDG and these should not be misinterpreted as lymphomatous BM involvement before or after therapy. Erroneous upstaging could expose the patient to longer, more toxic therapy that may not resolve the uptake on follow-up scans (ie, nonlymphomatous lesions not responsive to lymphoma therapy). Examples of benign lesions include healing bone (eg, fractures, postsurgery), osteomyelitis, degenerative disease, giant cell tumors, osteoid osteomas, chondroblastomas, fibrous dysplasia, enchondromas, nonossifying fibromas, fibrous cortical defects, cortical desmoids tumors, bone infarcts, Schmorl’s node, dental disease, radiation necrosis, Paget’s disease, and sarcoidosis in the bone (34–44). Malignant bone tumors generally have higher FDG uptake than benign
Figure 4. Representative images of a 53-year-old male with Stage IVA diffuse large B-cell lymphoma. (a) Positron emission tomography (PET)/ computed tomography (CT) scan for initial staging revealed numerous foci of intense FDG uptake throughout the pelvic bones with no corresponding CT abnormalities and a left presacral soft-tissue mass with intense FDG uptake (arrows). Iliac crest bone marrow (BM) biopsy was positive. (b) Posttherapy PET/CT scan after six cycles of rituximab-cyclophosphamide, doxorubicin, vincristine, prednisolone (R-CHOP) chemotherapy revealed complete resolution of the FDG uptake in the pelvic bones and presacral soft-tissue mass. Posttherapy iliac crest BM biopsy was negative. He was considered to be in complete remission by International Workshop Criteria+PET and PET+BM. He received intrathecal prophylaxis after chemotherapy and remained in a complete remission at 35 months after the pretherapy PET/CT scan.
1181
GOUDARZI ET AL
Academic Radiology, Vol 17, No 9, September 2010
Figure 5. Representative sagittal (left 3 columns) and transaxial images (right 2 columns) of a 54-year-old female with Stage IVB diffuse large B-cell lymphoma. (a) FDG positron emission tomography (PET)/computed tomography (CT) scan for initial staging revealed extensive FDG-avid lymphadenopathy (ie, para-aortic nodes, arrows) and numerous foci of FDG uptake within bone marrow (BM) despite a normal appearance of the bone on CT (ie, L3 vertebral body; crosshairs and arrowhead). Iliac crest BM biopsy was positive for non-Hodgkin’s lymphoma. (b) Midtherapy PET/CT scan after three cycles of rituximab-cyclophosphamide, doxorubicin, vincristine, prednisolone (R-CHOP) chemotherapy revealed complete resolution of all FDG-avid lymphadenopathy, but increased FDG uptake in several vertebral bodies was suspicious for residual active lymphoma (b, crosshair and arrowheads). Mid-therapy BM biopsy was negative. She was considered to be in complete remission by International Workshop Criteria+PET, but partial remission by PET+BM criteria. (c) Therapy was intensified to autologous stem cell transplant. Posttransplant FDG PET/CT scan was negative for active lymphoma. She progressed in lymph nodes ~3 months posttransplant on FDG PET/ CT scan and expired 12.3 months after the pretherapy scan. This supports PET+BM assessment as a robust predictor of outcome.
bone lesions, but the uptake can be indistinguishable in some cases (37). The characteristic findings of the benign bone lesions on the correlative CT scan and the overall pattern of response are often very helpful in avoiding misinterpretation as lymphomatous involvement. Similar to new lung lesions posttherapy (21), it is unlikely for new BM disease to appear in the setting of an otherwise CR, particularly if the BM was negative before therapy. The use of biopsy alone for evaluation of BM as recommended in the IWC+PET (20) may not be suitable for all patients. If a patient with BM involvement before therapy does not have a biopsy after therapy, this patient is considered to have only a PR, even if there is CR clinically or radiographically by CT or PET (20). Likewise, a patient with BM 1182
involvement before therapy is considered to have a CR if the PET or CT and BM biopsy are negative posttherapy, even if there is intense residual FDG uptake in the BM at a distant site from standard biopsy (n = 5, 19% of our cohort). This appears to be a major deficiency in these criteria as these patients with PET-positive BM who are negative on standard non–PET-directed biopsy appear to be at high risk of early progression. In addition, BM biopsy site after therapy may need to be tailored based on the location of residual foci of uptake in BM, particularly if standard iliac crest biopsy was negative before therapy. On the other hand, PET scanning alone should not be used or considered reliable to monitor response in known BM disease by biopsy that is not detected on imaging at baseline.
Academic Radiology, Vol 17, No 9, September 2010
ASSESSING MARROW RESPONSE WITH FDG-PET
Figure 6. Kaplan-Meier estimates of overall survival. Overall survival was longer for patients who achieved a complete response (CR) vs. partial response or progressive disease by positron emission tomography (PET)+bone marrow (BM), but not by International Workshop Criteria (IWC)+PET. PET+BM appeared a stronger predictor of outcome for survival based on CR vs. non-CR than was IWC+PET. R-CHOP, rituximab-cyclophosphamide, doxorubicin, vincristine, prednisolone; CHOP, cyclophosphamide, doxorubicin, vincristine, prednisolone; R-ICE, rituximab-ifosfamide, carboplatin, etoposide; ABVD, doxorubicin, bleomycin, vinblastine, dacarbazine; ABVD-PE, doxorubicin, bleomycin, vinblastine, dacarbazine-cisplatin, etoposide.
We compared response assessment by IWC+PET without and with a metabolic assessment of BM and demonstrate the PET+BM criterion adds information for a significant number of patients (5 of 27, 19%) with PET-positive BM disease during or after therapy. The SUL in BM of these patients declined, similar to the nodal disease, but not to background levels. For two patients who progressed at the sites of residual uptake in BM and subsequently died from NHL, the PET+BM criterion was a more accurate assessment of response than IWC+PET. In two cases in which PET+BM criterion was associated with a good outcome, therapy was intensified. It is unclear if these patients would have achieved disease-free courses had standard therapy alone been applied. These results need to be validated in larger, more homogeneous patient populations. Previous literature in aggressive lymphoma demonstrates lower complete response rates for patients with BM involvement compared to those without BM disease (45), but it is difficult to ascertain the precise locations of the residual disease. At the time of relapse, Schein et al found that after a complete remission most relapses (69%) occur in lymph nodes alone and no patient relapsed in a previously uninvolved extranodal site (46). In a study of 52 patients with HL (15% with extranodal disease at initial staging), 15% relapsed in extranodal sites, but again the exact sites of extranodal disease at diagnosis and relapse were not specified (47). FDG PET/
CT may be another tool for helping to further refine our understanding of patterns of relapse in patients with lymphoma, especially the evaluation of BM disease. The discordancy rate between biopsy and FDG PET for detecting lymphoma in BM in our study is higher (67%, 18 of 27 patients) compared to the published literature (~20%) (6,11–13,15,16). However, part of this variance may be due to our selection of PET-positive patients in the BM as the entry cohort. We specifically excluded patients with negative FDG PET scans in the BM, and although some may have been discordant with PET, many may also have had negative BM biopsies. A potential limitation of our study is its retrospective nature and heterogeneous patient population as regards lymphoma classification, treatment types, and timing of the PET studies after therapy. Although the performance of PET to detect macroscopic disease in BM may vary according to lymphoma subtype (12,17) and volume of disease, we did not consider this a significant limitation for the task we chose to evaluate in this study—does increased FDG uptake in LN and BM due to lymphoma behave similarly after treatment. All LN tumors had high levels of FDG uptake at baseline, including two with low-grade follicular types. No statistically significant differences in the results were observed when analyses were performed excluding patients with follicular lymphoma and those treated with radioimmunotherapy (data not shown). Obviously, PET scanning alone should not be used or considered reliable to 1183
GOUDARZI ET AL
Academic Radiology, Vol 17, No 9, September 2010
Figure 7. Representative sagittal images of a 26-year-old female with Stage IVB Hodgkin’s lymphoma before (a) and after (b) chemotherapy. Growth colony stimulating factors were administered with chemotherapy. (a) Pretherapy, intense FDG uptake is seen fusing to multiple vertebral bodies, most pronounced in T10 (crosshairs). Intense FDG uptake is also seen in lymph nodes behind the sternum. (b) Posttherapy, there is resolution of the FDG uptake in T10, which is now photopenic (crosshairs) compared to normal, stimulated bone marrow. The FDG uptake in the lymph nodes also resolved.
monitor response in BM disease not detected at baseline. Finally, measuring LNs on noncontrast CT scans can be difficult. The LNs evaluated in this study were chosen based on their suitability at baseline for accurate repeated measures and this was not considered a major limitation. In conclusion, anatomically ‘‘unmeasurable’’ BM infiltration with lymphoma often behaves similarly to LN disease after therapy and thus appears to be ‘‘measurable’’ by FDG PET/CT. FDG PET/CT is a valuable technique for monitoring tumor response in ‘‘measurable’’ disease (LN), and BM, which was previously considered to be ‘‘unmeasurable’’ 1184
by anatomical imaging. Additional prospective studies with more homogenous groups of patients and additional tumor types will be of interest to expand on, validate, and refine our results. Revision of response criteria to include PET of previously ‘‘unmeasurable’’ sites appears in order. The new PERCIST criteria (26) are a step in this direction. ACKNOWLEDGMENTS The authors would like to thank the editorial assistance of Mrs. Julia Buchanan.
Academic Radiology, Vol 17, No 9, September 2010
REFERENCES 1. Coller BS, Chabner BA, Gralnick HR. Frequencies and patterns of bone marrow involvement in non-Hodgkin’s lymphomas: observations on the value of bilateral biopsies. Am J Hematol 1977; 3:105–119. 2. Bitran JD, Golomb HM, Ultmann JE, et al. Non-Hodgkin’s lymphoma, poorly differentiated lymphocytic and mixed cell types: results of sequential staging procedures, response to therapy, and survival of 100 patients. Cancer 1978; 42:88–95. 3. 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 2000; 92: 205–216. 4. Tatsumi M, Cohade C, Nakamoto Y, et al. Direct comparison of FDG PET and CT findings in patients with lymphoma: initial experience. Radiology 2005; 237:1038–1045. 5. Nakamoto Y, Cohade C, Tatsumi M, et al. CT appearance of bone metastases detected with FDG PET as part of the same PET/CT examination. Radiology 2005; 237:627–634. 6. Carr R, Barrington SF, Madan B, et al. Detection of lymphoma in bone marrow by whole-body positron emission tomography. Blood 1998; 91: 3340–3346. 7. Pakos EE, Fotopoulos AD, Ioannidis JP. 18F-FDG PET for evaluation of bone marrow infiltration in staging of lymphoma: a meta-analysis. J Nucl Med 2005; 46:958–963. 8. Park YH, Kim S, Choi SJ, et al. Clinical impact of whole-body FDG-PET for evaluation of response and therapeutic decision-making of primary lymphoma of bone. Ann Oncol 2005; 16:1401–1402. 9. Park YH, Choi SJ, Ryoo BY, Kim HT. PET imaging with F-18 fluorodeoxyglucose for primary lymphoma of bone. Clin Nucl Med 2005; 30:131–134. 10. Moog F, Bangerter M, Diederichs C, et al. Extranodal malignant lymphoma: detection with F-l8-FDG PET versus CT. Radiology 1998; 206:475–481. 11. Moog F, Bangerter M, KotzerkeI, et al. FDG PET as a new approach to detect lymphomatous bone marrow. J Clin Oncol 1998; 16:603–609. 12. Elstrom R, Guan L, Baker G, et al. Utility of FDG-PET scanning in lymphoma by WHO classification. Blood 2003; 101:3875–3876. 13. Muslimani AA, Faraq HL, Francis S, et al. The utility of 18-F-fluorodeoxyglucose positron emission tomography in evaluation of bone marrow involvement by non-Hodgkin lymphoma. Am J Clin Oncol 2008; 31: 409–412. 14. Pelosi E, Pregno P, Penna D, et al. Role of whole-body [18F] fluorodeoxyglucose positron emission tomography/computed tomography (FDGPET/CT) and conventional techniques in the staging of patients with Hodgkin and aggressive non Hodgkin lymphoma. Radiol Med 2008; 113: 578–590. 15. Pelosi E, Penna D, Deandreis D, et al. FDG-PET in the detection of bone marrow disease in Hodgkin’s disease and aggressive non-Hodgkin’s lymphoma and its impact on clinical management. Q J Nucl Med Mol Imaging 2008; 52:9–16. 16. Schaefer NG, Strobel K, Taverna C, et al. Bone involvement in patients with lymphoma: the role of FDG-PET/CT. Eur J Nucl Med Mol Imaging 2007; 34: 60–67. 17. Kako S, Izutsu K, Ota Y, et al. FDG-PET in T-cell and NK-cell neoplasms. Ann Oncol 2007; 18:1685–1690. 18. Salaun PY, Gastinne T, Bodet-Milin C, et al. Analysis of 18F-FDG PET diffuse bone marrow uptake and splenic uptake in staging of Hodgkin’s lymphoma: a reflection of disease infiltration or just inflammation? Eur J Nucl Med Mol Imaging 2009; 36:1813–1821. 19. Cheson BD, Horning SJ, Coiffier B, et al. Report of an international workshop to standardize response criteria for non-Hodgkin’s lymphomas. NCI Sponsored International Working Group. J Clin Oncol 1999; 17: 1244–1253. 20. Cheson BD, Pfistner B, Juweid ME, et al. Revised response criteria for malignant lymphoma. J Clin Oncol 2007; 25:579–586. 21. Juweid ME, Stroobants S, Hoekstra OS, et al. Use of positron emission tomography for response assessment of lymphoma: consensus of the Imaging Subcommittee of International Harmonization Project in Lymphoma. J Clin Oncol 2007; 25:571–578. 22. Sugawara Y, Fisher SJ, Zasadny KR, et al. Preclinical and clinical studies of bone marrow uptake of fluorine-1-fluorodeoxyglucose with or without granulocyte colony-stimulating factor during chemotherapy. J Clin Oncol 1998; 16:173–180.
ASSESSING MARROW RESPONSE WITH FDG-PET
23. Jacene HA, Ishimori T, Engles JM, et al. Effects of pegfilgrastim on normal biodistribution of 18F-FDG: preclinical and clinical studies. J Nucl Med 2006; 47:950–956. 24. Blodgett TM, Ames JT, Torok FS, et al. Diffuse bone marrow uptake on whole-body F-18 fluorodeoxyglucose positron emission tomography in a patient taking recombinant erythropoietin. Clin Nucl Med 2004; 29: 161–163. 25. Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009; 45:228–247. 26. Wahl RL, Jacene H, Lodge MA. From RECIST to PERCIST: evolving Considerations for PET response criteria in solid tumors. J Nucl Med 2009; 50(Suppl 1). 122S–50S. 27. Freudenberg LS, Antoch G, Schu¨tt P, et al. FDG-PET/CT in re-staging of patients with lymphoma. Eur J Nucl Med Mol Imaging 2004; 31: 325–329. 28. Naumann R, Vaic A, Beuthien-Baumann B, et al. Prognostic value of positron emission tomography in the evaluation of post-treatment residual mass in patients with Hodgkin’s disease and non-Hodgkin’s lymphoma. Br J Haematol 2001; 115:793–800. 29. Lin C, Itti E, Haioun C, et al. Early 18F-FDG PET for prediction of prognosis in patients with diffuse large b-cell lymphoma: SUV-based assessment versus visual analysis. J Nucl Med 2007; 48:1626–1632. 30. Murata Y, Kubota K, Yukihiro M, et al. Correlations between 18F-FDG uptake by bone marrow and hematological parameters: measurements by PET/CT. Nucl Med Biol 2006; 33:999–1004. 31. Nahmias C, Hanna WT, Wahl LM, et al. Time course of early response to chemotherapy in non-small cell lung cancer patients with 18F-FDG PET/ CT. J Nucl Med 2007; 48:744–751. 32. Kazama T, Swanston N, Podoloff DA, et al. Effect of colony-stimulating factor and conventional- or high-dose chemotherapy on FDG uptake in bone marrow. Eur J Nucl Med Mol Imaging 2005; 32:1406–1411. 33. Lin EC. FDG PET/CT flip flop phenomenon in treated lymphoma of bone. Clin Nucl Med 2006; 31:803–805. 34. Shammas A, Lim R, Charron M. Pediatric FDG PET/CT: physiologic uptake, normal variants, and benign conditions. Radiographics 2009; 29: 1467–1486. 35. Mahmood S, Llano M. Paget disease of the humerus mimicking metastatic disease in a patient with metastatic malignant mesothelioma on whole body F-18 FDG PET/CT. Clin Nucl Med 2008; 33:510–512. 36. Bybel B. F-18 FDG PET/CT following dental extraction in a patient with head and neck cancer. Clin Nucl Med 2007; 32:741–742. 37. Shin D, Shon O, Han D, et al. The clinical efficacy of 18F-FDG-PET/CT in benign and malignant musculoskeletal tumors. Ann Nucl Med 2008; 22: 603–609. 38. Aoki J, Endo K, Watanabe H, et al. FDG-PET for evaluating musculoskeletal tumors. J Orthop Sci 2003; 8:435–441. 39. Feldman F, Heertum RV, Manos C. 18FDG PET scanning of benign and malignant musculoskeletal lesions. Skeletal Radiol 2003; 32:201–208. 40. Mester U, Even-Sapir E. Increased 18F-fluorodeoxyglucose uptake in benign, nonphysiologic lesions found on whole-body positron emission tomography/computed tomography (PET/CT): accumulated data from four years of experience with PET/CT. Semin Nucl Med 2007; 37: 206–222. 41. Dimitrakopoulou-Strauss A, Strauss LG, Heichel T, et al. The role of quantitative 18F-FDG PET studies for the differentiation of malignant and benign bone lesions. J Nucl Med 2002; 43:510–518. 42. Aoki J, Watanabe H, Shinozaki T, et al. FDG PET of primary benign and malignant bone tumors: standardized uptake value in 52 lesions. Radiology 2001; 219:774–777. 43. Grigolon MV, Delbeke D. F-18FDG uptake in a bone infarct: a case report. Clin Nucl Med 2001; 26:613–614. 44. Rosen R, Fayad L, Wahl RL. Increased 18F-FDG uptake in degenerative disease of the spine: characterization with 18F-FDG PET/CT. J Nucl Med 2006; 47:1274–1280. 45. A predictive model for aggressive non-Hodgkin’s lymphoma. The International Non-Hodgkin’s Lymphoma Prognostic Factors Project. N Engl J Med 1993; 329:987–994. 46. Schein P, Chabner B, Canellos G, et al. Non-Hodgkin’s lymphoma: patterns of relapse from complete remission after combination chemotherapy. Cancer 1975; 35:354–357. 47. Wirth A, Corry J, Laidlaw C, et al. Salvage radiotherapy for Hodgkin’s disease following chemotherapy failure. Int J Radiation Oncology Biol Phys 1997; 39:599–607.
1185
T
umor involvement of bone marrow (BM) in patients with lymphoma defines a higher stage and portends a worse prognosis. Biopsy is the current standard for diagnosing lymphomatous infiltration of BM. Sensitivity for identifying disease in BM increases 10%–20% by multiple sampling; however, even with negative bilateral biopsies, patients may subsequently prove to have BM disease (1,2). Bone and BM lesions are considered ‘‘nonmeasurable’’ in therapy response criteria, which are primarily based on anatomic imaging. The terms ‘‘measurable’’ and ‘‘nonmeasurable’’ were introduced in the Response Evaluation Criteria in Solid Tumors (RECIST) approach (3). Per RECIST, ‘‘measurable lesions’’ can be accurately measured in at least one dimension with the longest diameter being $10 or 20 mm depending on computed tomography (CT) technique.
Acad Radiol 2010; 17:1175–1185 From the Division of Nuclear Medicine, Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University, 601 North Caroline Street, JHOC 3223, Baltimore, MD 21287-0817 (B.G., H.A.J., R.L.W.). Received February 18, 2010; accepted May 1, 2010. Supported in part by a grant from NIH/NCI (3 P30 CA006973-43S2 Abeloff/Wahl for IRAT: Imaging Response Assessment Teams in Cancer Center Supplement). Address correspondence to: R.L.W. e-mail: [email protected] ªAUR, 2010 doi:10.1016/j.acra.2010.05.001
Bone lesions and several others (eg, ascites, pleural/pericardial effusions, leptomeningeal disease, lymphangitis cutis/pulmonis, cystic lesions) are specifically considered as ‘‘truly nonmeasurable’’ per RECIST. Although no direct measurements of tumor burden are taken, response of ‘‘nonmeasurable’’ lesions is ‘‘considered’’ when assigning a posttreatment response classification. Numerous studies have demonstrated that positron emission tomography (PET) with [18F]-2-fluoro-2-deoxy-Dglucose (FDG) detects BM tumor involvement in patients with lymphoma, often in the presence of normal CT scans (4,5), and provides complementary information to BM biopsy (6–17). The discordancy rate between FDG PET and BM biopsy is ~20% (6,11–13,15–17). False-negative PET scans for BM disease occur with lower tumor burden (<35%) in BM, low-level FDG uptake in the primary nodal disease, or specific histologic subtype (12,17). Positive PET scans in BM with negative BM biopsy (ie, false-positive PET scans) have also been reported. This may be due to sampling error, and disease is often detected with PETdirected rebiopsy of focal FDG uptake distant from the site of the initial negative BM biopsy (6,11,16). In early studies, diffuse FDG uptake throughout the BM was another cause of false-positive PET scans. At least in Hodgkin’s lymphoma (HL), it is now recognized that diffuse, 1175
GOUDARZI ET AL
TABLE 1. Patient Characteristics (Total Group, n = 27) Characteristics Sex (M/F) 20/7 Age (mean SD) 45 21 years Histologies HL 9 Nodular sclerosing 7 Mixed cellularity 1 Not otherwise specified 1 NHL 18 Large B cell 16* Follicular 2 Pretherapy PET indication Initial staging 18 Restaging 9 Interval therapy 21 Chemotherapy aloney Radioimmunotherapy 4 Donor lymphocyte infusion 1 Stem cell transplant 1 Timing of posttherapy PET Mid-therapy After 2 cycles of chemotherapy 4 After 3 cycles of chemotherapy 10 13 After completion of therapyz Days between (mean SD) Pre- and posttherapy PET 95 50 Last therapy day and posttherapy PET 31 27 Last therapy day and pretherapy, restaging PET 100 54 HL, Hodgkin’s lymphoma; NHL, non-Hodgkin’s lymphoma; PET, positron emission tomography. *Including 2 transformed from low-grade follicular NHL. y Chemotherapy regimens consisted of R-CHOP in 12 and R-ICE in 1 with diffuse large B-cell NHL. The patient with Burkitt’s NHL received CHOP. HL regimens included ABVD (n = 2); ABVE-PC (n = 2); BCVD, gemcitabine/cisplatin in 1 each and 1 unknown regimen. z n = 7 after chemotherapy; n = 4 after radioimmunotherapy; n = 1 each after donor lymphocyte infusion and stem cell transplant.
homogeneous BM uptake at initial staging may be due to tumor infiltration but is more likely the result of reactive changes (6,18). The widely accepted International Workshop Criteria (IWC) (19) for assessing response of non-Hodgkin’s lymphoma (NHL) to treatment were revised to include an assessment of changes in tumor metabolism using FDG PET after treatment (IWC+PET) (20). In IWC+PET, biopsy remains the standard test for detecting lymphoma in BM before and after therapy (20). This is despite recommendations that clearly increased (multi)-focal FDG uptake in BM be interpreted as positive for lymphoma in companion guidelines for interpreting PET scans (21). One potential limitation of PET for evaluation of the BM in the posttherapy setting is that diffuse FDG uptake in reconstituting BM from endogenous or exogenous hematopoietic growth factors may obscure residual focal lesions (22–24). 1176
Academic Radiology, Vol 17, No 9, September 2010
Most previous investigations evaluated FDG uptake in BM lesions qualitatively and focused on disease detection and concordance with biopsy. This retrospective observational study was performed to compare the ability of FDG PET/ CT as a noninvasive imaging method to determine treatment response in anatomically ‘‘nonmeasurable’’ lesions (BM) versus measurable lesions (lymph nodes) by semiquantitative analysis and to determine the effect of including a metabolic evaluation of BM in the IWC+PET response assessment on overall survival. We hypothesized that BM involvement with tumor would be detectable, metabolically measurable with the standardized uptake value (SUV), would respond in a similar manner to anatomically ‘‘measurable’’ nodal disease, and would provide valuable prognostic information beyond that routinely available.
MATERIALS AND METHODS Our institutional review board approved this study under an expedited review and waived the requirement for informed consent. Patient Selection
PET/CT reports from 1100 patients with lymphoma imaged between June 2003 and April 2006 were retrospectively reviewed. Twenty-seven met the following inclusion criteria and were selected as the study population: diagnosis of HL or NHL; unifocal or multifocal FDG-avid lesions reported in BM (with or without cortical involvement); $1 involved, anatomically measurable lymph node (LN) or mass lesion detected on pretherapy PET/CT scan obtained; mid- or posttherapy PET/CT scan. Eighteen patients presented at initial diagnosis, whereas nine were studied at the time of recurrence or persistent disease because a restaging scan was required for possible change in management. Patient characteristics and interval therapies are shown in Table 1. PET/CT Imaging
Patients were fasting at least 4 hours and had serum glucose levels <200 mg/dL before intravenous injection of a weight-based dose of FDG (0.22 mCi/kg, 8.14 MBq/kg). Oral contrast (1.3% Readi-cat, EZEM, Lake Success, NY) was administered. After an approximately 60-minute tracer uptake phase, a combined PET/CT scan (Discovery LS or ST; GE Healthcare, Waukesha, WI) was obtained from mid-skull to midfemur level or beyond if necessary. Whole-body CT, without intravenous contrast, was performed first, with 4- or 16-slice multidetector helical scanners. CT parameters were specific for the particular scanner, but are summarized for the Discovery ST and LS: 120–140 kVp, weight-based amperage (20–250 mA), 0.5 and 0.8 seconds per CT rotation, pitch of 0.984:1 and 1.5:1, and 3.75 and 5 mm reconstructed slice
Academic Radiology, Vol 17, No 9, September 2010
thickness, respectively. Emission data were acquired in the two-dimensional mode for 5 minutes per bed position and were reconstructed using the ordered-subset expectation maximization algorithm (two iterations, 21 or 28 subsets), Gaussian post- filter with a 128 128 matrix, and CT attenuation correction. Posttherapy PET/CT scans were performed midchemotherapy, after two (n = 4) or three (n = 10) cycles in 14 patients, and after the completion of therapy in 13 patients (n = 7 chemotherapy; n = 4 radioimmunotherapy; n = 1 donor lymphocyte infusion; n = 1 myeloablative stem cell transplant). Times between pre- and posttherapy scans and last treatment and posttherapy scans are listed in Table 1. Image Interpretation
Two nuclear medicine physicians with experience in PET/ CT reviewed the original images for the 27 patients meeting study criteria, first independently and then in consensus. PET, CT, and fused PET/CT images were viewed on a single display screen using Xeleris software (GE Healthcare, Waukesha, WI). On the CT part of the pretherapy PET/CT images, up to six FDG-avid LNs were selected on the basis of their size (lesions with the longest diameters) and their suitability for accurate repeated measurements by imaging techniques. One hundred and thirty LNs were located in the following regions: neck (n = 16), axilla (n = 20), para-aortic/aortocaval (n = 13), mediastinum (n = 16), supraclavicular (n = 3), subpectoral (n = 3), hilum (n = 3), paravertebral (n = 2), pelvic (n = 13), inguinal (n = 7), upper abdomen (n = 13), and other (n = 21). The product of the long and short axis diameters for each LN was calculated and baseline sum of products of the diameters (SPD) determined. The same LNs were measured on the CT part of the posttherapy PET/CT scan and posttherapy SPD calculated. Large LN masses were included in the analysis if they had definitive borders. Liver, spleen, and lung masses were excluded. For the metabolic semiquantitative analysis, a 1-cm circular region of interest (ROI) was drawn around the most intense area of FDG accumulation in the same LNs or LN masses measured on CT. The mean and maximum (max) standardized uptake values corrected for lean body mass (SUL) within the 1-cm ROI were determined for each LN pre- and posttherapy. If a LN completely resolved on the posttherapy scan, background SUL was determined in the region of previous tumor. SULmean and SULmax were determined in up to six different foci of increased FDG uptake in BM. If the focus of FDG uptake completely resolved on the posttherapy scan, background SUL was determined in the region of the prior focus. Average Hounsfield units in a 1-cm diameter ROI were determined on CT corresponding to the location of the foci of increased FDG in BM lesions before and after therapy. Normal BM metabolic uptake was determined in the lumbar vertebrae by drawing a 1-cm diameter ROI.
ASSESSING MARROW RESPONSE WITH FDG-PET
It was not possible to histologically confirm all areas of increased FDG uptake as BM disease, but lesions were chosen for evaluation considering the overall pattern of disease on the baseline PET and correlative anatomic imaging. The results regarding lymphomatous infiltration of the BM on FDG PET/CT scanning were compared with biopsy and histology when available or other imaging modalities, such as magnetic resonance imaging (MRI), CT, or bone scan. Posttherapy Response Assessments
Posttherapy response was first classified according to IWC+PET (20). A complete response (CR) required a negative FDG PET scan and a negative BM biopsy (if positive pretherapy). For a partial response (PR), SPD of LNs must have declined by $50% and at least one LN remained PET positive posttherapy by qualitative visual assessment. In the setting of PET positivity in LNs posttherapy, BM biopsy results were ‘‘irrelevant’’ for response classification (ie, a PR would have been assigned if the BM biopsy was positive or negative) (20). Progressive disease (PD) was defined as $50% increase in SPD or size of one LN or any new FDG-avid lesion that was confirmed by histology or other modalities. New foci of FDG uptake in BM was not considered PD because BM biopsy is the standard procedure for assessment of BM and for all new previously uninvolved sites, histologic confirmation is necessary (19,20). IWC+PET response was then reclassified adding a qualitative visual assessment of metabolic response of BM lesions after therapy (PET+BM). In addition to a CR by IWC+PET, a CR by PET+BM required FDG uptake in BM lesions to be less than baseline and less than or equal to background BM. A CR or PR by IWC+PETwas considered a PR by PET+BM if there was residual focal FDG uptake in BM lesions that was less than baseline but greater than or equal to background BM. PD by PET+BM was defined as for IWC+PET but also as present if the intensity of an FDG-avid focus in BM visually increased or a new BM focus of FDG uptake consistent with a lymphomatous pattern was seen. Statistical Analyses
Statistical analyses were performed using S-Plus (Insight Corporation, Seattle, WA), Systat (Cranes Software, Karnataka, India), and/or Microsoft Excel (Microsoft, Seattle, WA). Differences of means were compared using two-tailed paired t-tests and correlations explored using Pearson correlation coefficients and z-tests. P values # .05 were considered statistically significant. Agreement between IWC+PET and PET+BM response assessments was evaluated with the kappa statistic (k). Overall survival (OS) was defined as time between pretherapy FDG PET/CT scan and death from any cause. OS was correlated with response using Kaplan-Meier curves and log-rank tests. 1177
GOUDARZI ET AL
Academic Radiology, Vol 17, No 9, September 2010
TABLE 2. Lymph Node Size and Lymph Node and Bone Marrow FDG Uptake Before Therapy After Therapy P Value Lymph node (n = 130) SPD (size, mm2) SULmean SULmax Bone marrow lymphoma (n = 91) SULmean SULmax Normal bone marrow SULmean SULmax
2337 1810 6.94 4.96 8.74 6.19
309 323 1.02 1.00 1.29 1.32
<.001 <.001 <.001
6.81 4.58 8.26 5.2
1.84 1.58 2.16 1.95
<.001 <.001
0.85 0.33 1.00 0.34
0.93 0.32 1.12 0.37
.28 .37
SPD, sum of products of the diameters; SULmean, mean standardized uptake value; SULmax, maximum standardized uptake value.
RESULTS Mean SULmean of LNs and LN size were significantly higher pre- versus posttherapy (6.94 4.96 vs. 1.02 1.00 and 2337 mm2 1810 vs. 309 mm2 323, P < .001, respectively; Table 2). SULmean of BM foci decreased significantly posttherapy (6.81 4.58 to 1.84 1.58, P < .001). An example is shown in Figures 1A and 1B. Similar results were obtained analyzing SULmax (Table 2). Normal BM metabolic activity in the vertebrae was unchanged with treatment (Table 2). No significant differences were found in HU of BM lesions before and after therapy (data not shown). Strong positive correlations were observed between percentage decreases of LN SPD and LN SULmean after therapy (r = 0.84, P < .001) and LN SULmean and BM SULmean (r = 0.76, P < .001) (Figs 2a,b). There was a positive correlation between percentage decrease of LN SPD and BM SULmean after therapy (r = 0.56, P = .002) (Fig 2c). Similar strong correlations were observed when evaluating percentage decreases of SULmax of the LN or BM and LN SPD (LN SPD vs. LN SULmax: r = 0.84, P < .0001; LN SULmax vs. BM SULmax: r = 0.79, P < .0001; LN SPD vs. BM SULmax: r = 0.59, P = .001). Comparing pre- versus posttherapy scans, no significant differences were found in several patient and technical parameters known to affect SUL, including blood glucose levels (95 13 vs. 99 13 mg/dL, P = .22), FDG uptake phase time (64 10 vs. 66 13 minutes, P = .45), and lean body mass (59 10 vs. 59 11 kg, P = .11). A summary of histologic and imaging correlation/confirmation is presented in Figure 3. Twenty-four patients had standard BM aspirations and biopsies performed at the iliac crest at a median of 2 days before the PET/CT scan (range, 156 days before to 63 days after the PET/CT scan). Four patients had positive biopsies for lymphomatous involvement in BM at the iliac crest. In 11 patients with a positive FDG PET/CT in BM but a negative iliac crest biopsy, contempo1178
raneous studies (range 14 days before PET/CT to 39 days after, median 0 days) revealed BM infiltration at the site of abnormal FDG uptake on PET as follows: one with PETdirected BM biopsy, one with biopsy before PET based on x-ray, one with bone scan and CT scans, three with bone scans, two with CT, and three with MRI. In the other nine patients, further imaging studies were suggestive of BM involvement in four and BM involvement was not confirmed in five. Directed BM biopsy was the only site of biopsy in 3 of 27 patients. In these cases, preceding imaging studies suggested bone/BM abnormalities and biopsies were performed before FDG PET/CT imaging. In all three cases, there was intense FDG uptake corresponding to abnormal CT findings on bone windows, greater than would be expected from biopsy alone. Overall, nine patients had a positive BM biopsy pretherapy. BM biopsy was repeated after therapy in six of these patients and all were negative. Nineteen of 27 (70%) patients achieved a CR after therapy, 5 (18%) had a PR, and 3 (12%) had PD by IWC+PET. IWC+PET and PET+BM response assessments (Table 3) agreed in 22 of 27 cases (81%, k = 0.67; Fig 4). In five patients (19%), response was classified as CR by IWC+PET but PR by PET+BM (Fig 5). Four of these five response assessments were made on mid-therapy posttreatment scans and one was based on posttransplant FDG PET/CT scan. For these five patients, three had iliac crest biopsies only before therapy. Median follow-up time was 30.2 months (range 4.1–77.8 months). OS was longer for patients who achieved a CR by IWC+PET and PET+BM (Fig 6). Of the five patients reclassified from CR to PR by PET+BM, two progressed at sites of residual FDG uptake in BM and died of progressive NHL 26.2 and 20.3 months after pretherapy PET. One achieved a CR after continuing with standard chemotherapy and was disease-free 19.1 months after pretherapy PET. Two had therapy intensified mid-treatment to stem cell transplant and converted to CR on posttherapy PET. One relapsed in lymph nodes 3 months posttransplant and died 12.3 months after the pretherapy scan. The other received additional posttransplant external beam radiation therapy to the BM lesion and was disease free 44.1 months after the pretherapy scan. Based on the available clinical history, 11 of 27 patients received hematopoietic growth factors a median of 12 days (range 3–43 days) before posttherapy PET. Of these 11, 5 were felt to visually have diffusely increased uptake in BM from the growth factors without any definite foci of more intense uptake to suggest residual BM disease. Eleven patients did not receive hematopoietic growth factors and receipt of growth factors is unknown in five patients, although one patient’s scan suggested recent receipt of growth factors. Three of the five patients with residual foci of uptake in BM posttherapy did not receive hematopoietic growth factors before posttherapy PET, one was unknown, and one had focal BM uptake above background suggesting residual disease in BM despite receiving growth factors 11 days before midtherapy PET scan (Fig 5).
Academic Radiology, Vol 17, No 9, September 2010
ASSESSING MARROW RESPONSE WITH FDG-PET
Figure 1. (a) Lymph node involvement before and after therapy is shown on these scans (computed tomography [CT], left; positron emission tomography [PET]/CT, middle; PET, right). Before therapy, the lymph node (arrows) measured 28 16 mm and had a mean standardized uptake value (SULmean) of 8.11. After therapy, the lymph node (crosshairs) measured 16 8 mm and had a SULmean of 0.72. (b) Bone marrow involvement in the sternum (arrowheads) is shown for the same patient (CT, left; PET/CT, middle; PET, right). Before therapy, SULmean in the BM lesion was 6.76 and after therapy it was 0.84.
1179
GOUDARZI ET AL
Academic Radiology, Vol 17, No 9, September 2010
Figure 2. Correlations between percent change of anatomic and metabolic parameters. There are strong positive correlations between (a) percent change in sum of products of the diameters (SPD) of LN and percent change in lymph node (LN) mean standardized uptake value (SULmean) after therapy (r = 0.84, P < .001), and (b) percent change in LN SULmean and bone marrow (BM) SULmean after therapy (r = 0.76, P < .001). (c) There is a positive correlation between the percent change in LN SPD and BM SULmean after therapy (r = 0.56, P = .002).
DISCUSSION Bone lesions have been considered ‘‘truly nonmeasurable’’ in criteria for monitoring response of tumor to therapy (RECIST and IWC). Similar to IWC, the RECIST guideline was recently revised (version 1.1); in RECIST 1.1, special consideration is given to bone lesions (25). The identifiable soft-tissue components of lytic bones lesions may be considered measurable if they meet all other criteria for measurable disease and can be evaluated by CT or MRI. PET scan is specifically considered an inadequate imaging technique to measure bone lesions. The results of our report suggest otherwise. For this investigation, we chose to evaluate the use of FDG PET for monitoring the response of BM disease in patients with lymphoma because this population relatively frequently has BM disease, has disease that responds to
Figure 3. Flow chart of histologic and imaging correlation/confirmation of FDG-avid bone marrow (BM) lesions. PET, positron emission tomography; CT, computed tomography; MRI, magnetic resonance imaging.
1180
therapy, and is now routinely assessed with FDG PET before and after therapy. Our discussion primarily focuses on PET for monitoring response of BM disease in patients with lymphoma, but the concepts should be considered for broader applicability to other tumor types and response criteria beyond the IWC for lymphoma. The recently proposed PET response criteria in solid tumors (PERCIST) 1.0 criteria allow for inclusion of BM disease in the treatment response assessment, just like other foci of FDGavid disease (26). In our study population, we successfully measured FDG uptake in regions of BM apparently infiltrated with lymphoma before and after therapy using the SUL parameter. The data presented support our hypotheses that changes in metabolic activity of lymphoma in the BM are in the same direction and magnitude as in nodal disease. To the best of our knowledge, no previous systematic studies have demonstrated this relationship offering a possible explanation for the lack of inclusion of assessment of BM disease in previous PET-based response criteria. The significance of semiquantitative uptake determinations in nodal lymphoma remains controversial. Even more limited data exist regarding the use of semiquantitative analyses of FDG uptake for foci of lymphomatous infiltration of BM. Schaefer et al reported a mean SUV of 6.26 3.22 in BM lesions (16) and our results are similar. Posttherapy SULmean of BM lesions decreased to 1.84 1.58 in our study. Schaefer et al did not compare pre- and posttherapy SUVs (16). Some authors have attempted to define ‘‘cutoff ’’ values for diagnosing residual disease or predicting prognosis in lymphoma (27,28). A sensitivity of 86% and specificity of 100% for the detection of residual/recurrent lymphoma by PET/CT was reported using a cutoff SUVmax of 2.5 (27). In 58 patients with HL or NHL who achieved a CR unconfirmed after therapy, FDG PET had a high accuracy for predicting early recurrence using a threshold SUVmax of 3 for the presence of recurrent lymphoma (28). Given the variability in cutoff values for level of FDG uptake, we did not attempt to define such cutoff values for LN or BM lesions in our patient population.
Academic Radiology, Vol 17, No 9, September 2010
TABLE 3. Responses After Therapy
IWC, International Workshop Criteria; PET, positron emission tomography; BM, bone marrow; CR, complete response; PR, partial response.
The IWC+PET response criteria rely on a visual qualitative assessment of changes in FDG uptake between baseline and posttherapy scans (20). Any additional benefit of semiquantitative analyses for characterizing response to therapy has yet to be definitively demonstrated, but preliminary data suggest this to be true (29). We categorized patients’ response to therapy according to IWC+PET criteria in this study, but we did not present these results versus changes in BM uptake because the majority of our patients had CR (70%) after therapy. In this study, each patient was used as his or her own control and we compared pre- and posttherapy SUL and also correlated percent changes rather than crude values. Unlike chemotherapy alone, hematopoietic growth factors cause a diffuse increase in FDG uptake in normal BM (22– 24,30–32), most markedly near the time of administration. The uptake declines, but can remain above baseline for 4 weeks after short-acting growth factors (22). A time course
ASSESSING MARROW RESPONSE WITH FDG-PET
of uptake after long-acting growth factors is not wellestablished. This is problematic because residual focal lesions in the BM may not only be obscured by the diffuse uptake but the differences in time between hematopoietic growth factor administration and PET can result in heterogeneity in FDG uptake in the BM between patients. In this study, diffuse FDG uptake in BM resulting from hematopoietic growth factors was noted in 6 of 27 patients posttheraphy, 5 with FDG uptake less than background BM in the locations of BM disease. Five achieved a CR in their nodal disease and, at last follow-up, median survival was 44 months. The sixth patient achieved a PR in nodal disease and died 13 months after the PET scan. We caution that application of a metabolic assessment of the BM (qualitative or quantitative) is limited in the setting of hematopoietic growth factors and may depend on the timing of the posttherapy PET scan after hematopoietic growth factor administration. However, we observed a ‘‘flip-flop’’ phenomenon (33), photopenia in the location of previous disease compared to stimulated normal BM in two cases (Fig 7), which was reassuring for the absence of metabolically active disease in BM. Primary malignant bone tumors and many other benign bone lesions have been reported to accumulate FDG and these should not be misinterpreted as lymphomatous BM involvement before or after therapy. Erroneous upstaging could expose the patient to longer, more toxic therapy that may not resolve the uptake on follow-up scans (ie, nonlymphomatous lesions not responsive to lymphoma therapy). Examples of benign lesions include healing bone (eg, fractures, postsurgery), osteomyelitis, degenerative disease, giant cell tumors, osteoid osteomas, chondroblastomas, fibrous dysplasia, enchondromas, nonossifying fibromas, fibrous cortical defects, cortical desmoids tumors, bone infarcts, Schmorl’s node, dental disease, radiation necrosis, Paget’s disease, and sarcoidosis in the bone (34–44). Malignant bone tumors generally have higher FDG uptake than benign
Figure 4. Representative images of a 53-year-old male with Stage IVA diffuse large B-cell lymphoma. (a) Positron emission tomography (PET)/ computed tomography (CT) scan for initial staging revealed numerous foci of intense FDG uptake throughout the pelvic bones with no corresponding CT abnormalities and a left presacral soft-tissue mass with intense FDG uptake (arrows). Iliac crest bone marrow (BM) biopsy was positive. (b) Posttherapy PET/CT scan after six cycles of rituximab-cyclophosphamide, doxorubicin, vincristine, prednisolone (R-CHOP) chemotherapy revealed complete resolution of the FDG uptake in the pelvic bones and presacral soft-tissue mass. Posttherapy iliac crest BM biopsy was negative. He was considered to be in complete remission by International Workshop Criteria+PET and PET+BM. He received intrathecal prophylaxis after chemotherapy and remained in a complete remission at 35 months after the pretherapy PET/CT scan.
1181
GOUDARZI ET AL
Academic Radiology, Vol 17, No 9, September 2010
Figure 5. Representative sagittal (left 3 columns) and transaxial images (right 2 columns) of a 54-year-old female with Stage IVB diffuse large B-cell lymphoma. (a) FDG positron emission tomography (PET)/computed tomography (CT) scan for initial staging revealed extensive FDG-avid lymphadenopathy (ie, para-aortic nodes, arrows) and numerous foci of FDG uptake within bone marrow (BM) despite a normal appearance of the bone on CT (ie, L3 vertebral body; crosshairs and arrowhead). Iliac crest BM biopsy was positive for non-Hodgkin’s lymphoma. (b) Midtherapy PET/CT scan after three cycles of rituximab-cyclophosphamide, doxorubicin, vincristine, prednisolone (R-CHOP) chemotherapy revealed complete resolution of all FDG-avid lymphadenopathy, but increased FDG uptake in several vertebral bodies was suspicious for residual active lymphoma (b, crosshair and arrowheads). Mid-therapy BM biopsy was negative. She was considered to be in complete remission by International Workshop Criteria+PET, but partial remission by PET+BM criteria. (c) Therapy was intensified to autologous stem cell transplant. Posttransplant FDG PET/CT scan was negative for active lymphoma. She progressed in lymph nodes ~3 months posttransplant on FDG PET/ CT scan and expired 12.3 months after the pretherapy scan. This supports PET+BM assessment as a robust predictor of outcome.
bone lesions, but the uptake can be indistinguishable in some cases (37). The characteristic findings of the benign bone lesions on the correlative CT scan and the overall pattern of response are often very helpful in avoiding misinterpretation as lymphomatous involvement. Similar to new lung lesions posttherapy (21), it is unlikely for new BM disease to appear in the setting of an otherwise CR, particularly if the BM was negative before therapy. The use of biopsy alone for evaluation of BM as recommended in the IWC+PET (20) may not be suitable for all patients. If a patient with BM involvement before therapy does not have a biopsy after therapy, this patient is considered to have only a PR, even if there is CR clinically or radiographically by CT or PET (20). Likewise, a patient with BM 1182
involvement before therapy is considered to have a CR if the PET or CT and BM biopsy are negative posttherapy, even if there is intense residual FDG uptake in the BM at a distant site from standard biopsy (n = 5, 19% of our cohort). This appears to be a major deficiency in these criteria as these patients with PET-positive BM who are negative on standard non–PET-directed biopsy appear to be at high risk of early progression. In addition, BM biopsy site after therapy may need to be tailored based on the location of residual foci of uptake in BM, particularly if standard iliac crest biopsy was negative before therapy. On the other hand, PET scanning alone should not be used or considered reliable to monitor response in known BM disease by biopsy that is not detected on imaging at baseline.
Academic Radiology, Vol 17, No 9, September 2010
ASSESSING MARROW RESPONSE WITH FDG-PET
Figure 6. Kaplan-Meier estimates of overall survival. Overall survival was longer for patients who achieved a complete response (CR) vs. partial response or progressive disease by positron emission tomography (PET)+bone marrow (BM), but not by International Workshop Criteria (IWC)+PET. PET+BM appeared a stronger predictor of outcome for survival based on CR vs. non-CR than was IWC+PET. R-CHOP, rituximab-cyclophosphamide, doxorubicin, vincristine, prednisolone; CHOP, cyclophosphamide, doxorubicin, vincristine, prednisolone; R-ICE, rituximab-ifosfamide, carboplatin, etoposide; ABVD, doxorubicin, bleomycin, vinblastine, dacarbazine; ABVD-PE, doxorubicin, bleomycin, vinblastine, dacarbazine-cisplatin, etoposide.
We compared response assessment by IWC+PET without and with a metabolic assessment of BM and demonstrate the PET+BM criterion adds information for a significant number of patients (5 of 27, 19%) with PET-positive BM disease during or after therapy. The SUL in BM of these patients declined, similar to the nodal disease, but not to background levels. For two patients who progressed at the sites of residual uptake in BM and subsequently died from NHL, the PET+BM criterion was a more accurate assessment of response than IWC+PET. In two cases in which PET+BM criterion was associated with a good outcome, therapy was intensified. It is unclear if these patients would have achieved disease-free courses had standard therapy alone been applied. These results need to be validated in larger, more homogeneous patient populations. Previous literature in aggressive lymphoma demonstrates lower complete response rates for patients with BM involvement compared to those without BM disease (45), but it is difficult to ascertain the precise locations of the residual disease. At the time of relapse, Schein et al found that after a complete remission most relapses (69%) occur in lymph nodes alone and no patient relapsed in a previously uninvolved extranodal site (46). In a study of 52 patients with HL (15% with extranodal disease at initial staging), 15% relapsed in extranodal sites, but again the exact sites of extranodal disease at diagnosis and relapse were not specified (47). FDG PET/
CT may be another tool for helping to further refine our understanding of patterns of relapse in patients with lymphoma, especially the evaluation of BM disease. The discordancy rate between biopsy and FDG PET for detecting lymphoma in BM in our study is higher (67%, 18 of 27 patients) compared to the published literature (~20%) (6,11–13,15,16). However, part of this variance may be due to our selection of PET-positive patients in the BM as the entry cohort. We specifically excluded patients with negative FDG PET scans in the BM, and although some may have been discordant with PET, many may also have had negative BM biopsies. A potential limitation of our study is its retrospective nature and heterogeneous patient population as regards lymphoma classification, treatment types, and timing of the PET studies after therapy. Although the performance of PET to detect macroscopic disease in BM may vary according to lymphoma subtype (12,17) and volume of disease, we did not consider this a significant limitation for the task we chose to evaluate in this study—does increased FDG uptake in LN and BM due to lymphoma behave similarly after treatment. All LN tumors had high levels of FDG uptake at baseline, including two with low-grade follicular types. No statistically significant differences in the results were observed when analyses were performed excluding patients with follicular lymphoma and those treated with radioimmunotherapy (data not shown). Obviously, PET scanning alone should not be used or considered reliable to 1183
GOUDARZI ET AL
Academic Radiology, Vol 17, No 9, September 2010
Figure 7. Representative sagittal images of a 26-year-old female with Stage IVB Hodgkin’s lymphoma before (a) and after (b) chemotherapy. Growth colony stimulating factors were administered with chemotherapy. (a) Pretherapy, intense FDG uptake is seen fusing to multiple vertebral bodies, most pronounced in T10 (crosshairs). Intense FDG uptake is also seen in lymph nodes behind the sternum. (b) Posttherapy, there is resolution of the FDG uptake in T10, which is now photopenic (crosshairs) compared to normal, stimulated bone marrow. The FDG uptake in the lymph nodes also resolved.
monitor response in BM disease not detected at baseline. Finally, measuring LNs on noncontrast CT scans can be difficult. The LNs evaluated in this study were chosen based on their suitability at baseline for accurate repeated measures and this was not considered a major limitation. In conclusion, anatomically ‘‘unmeasurable’’ BM infiltration with lymphoma often behaves similarly to LN disease after therapy and thus appears to be ‘‘measurable’’ by FDG PET/CT. FDG PET/CT is a valuable technique for monitoring tumor response in ‘‘measurable’’ disease (LN), and BM, which was previously considered to be ‘‘unmeasurable’’ 1184
by anatomical imaging. Additional prospective studies with more homogenous groups of patients and additional tumor types will be of interest to expand on, validate, and refine our results. Revision of response criteria to include PET of previously ‘‘unmeasurable’’ sites appears in order. The new PERCIST criteria (26) are a step in this direction. ACKNOWLEDGMENTS The authors would like to thank the editorial assistance of Mrs. Julia Buchanan.
Academic Radiology, Vol 17, No 9, September 2010
REFERENCES 1. Coller BS, Chabner BA, Gralnick HR. Frequencies and patterns of bone marrow involvement in non-Hodgkin’s lymphomas: observations on the value of bilateral biopsies. Am J Hematol 1977; 3:105–119. 2. Bitran JD, Golomb HM, Ultmann JE, et al. Non-Hodgkin’s lymphoma, poorly differentiated lymphocytic and mixed cell types: results of sequential staging procedures, response to therapy, and survival of 100 patients. Cancer 1978; 42:88–95. 3. 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 2000; 92: 205–216. 4. Tatsumi M, Cohade C, Nakamoto Y, et al. Direct comparison of FDG PET and CT findings in patients with lymphoma: initial experience. Radiology 2005; 237:1038–1045. 5. Nakamoto Y, Cohade C, Tatsumi M, et al. CT appearance of bone metastases detected with FDG PET as part of the same PET/CT examination. Radiology 2005; 237:627–634. 6. Carr R, Barrington SF, Madan B, et al. Detection of lymphoma in bone marrow by whole-body positron emission tomography. Blood 1998; 91: 3340–3346. 7. Pakos EE, Fotopoulos AD, Ioannidis JP. 18F-FDG PET for evaluation of bone marrow infiltration in staging of lymphoma: a meta-analysis. J Nucl Med 2005; 46:958–963. 8. Park YH, Kim S, Choi SJ, et al. Clinical impact of whole-body FDG-PET for evaluation of response and therapeutic decision-making of primary lymphoma of bone. Ann Oncol 2005; 16:1401–1402. 9. Park YH, Choi SJ, Ryoo BY, Kim HT. PET imaging with F-18 fluorodeoxyglucose for primary lymphoma of bone. Clin Nucl Med 2005; 30:131–134. 10. Moog F, Bangerter M, Diederichs C, et al. Extranodal malignant lymphoma: detection with F-l8-FDG PET versus CT. Radiology 1998; 206:475–481. 11. Moog F, Bangerter M, KotzerkeI, et al. FDG PET as a new approach to detect lymphomatous bone marrow. J Clin Oncol 1998; 16:603–609. 12. Elstrom R, Guan L, Baker G, et al. Utility of FDG-PET scanning in lymphoma by WHO classification. Blood 2003; 101:3875–3876. 13. Muslimani AA, Faraq HL, Francis S, et al. The utility of 18-F-fluorodeoxyglucose positron emission tomography in evaluation of bone marrow involvement by non-Hodgkin lymphoma. Am J Clin Oncol 2008; 31: 409–412. 14. Pelosi E, Pregno P, Penna D, et al. Role of whole-body [18F] fluorodeoxyglucose positron emission tomography/computed tomography (FDGPET/CT) and conventional techniques in the staging of patients with Hodgkin and aggressive non Hodgkin lymphoma. Radiol Med 2008; 113: 578–590. 15. Pelosi E, Penna D, Deandreis D, et al. FDG-PET in the detection of bone marrow disease in Hodgkin’s disease and aggressive non-Hodgkin’s lymphoma and its impact on clinical management. Q J Nucl Med Mol Imaging 2008; 52:9–16. 16. Schaefer NG, Strobel K, Taverna C, et al. Bone involvement in patients with lymphoma: the role of FDG-PET/CT. Eur J Nucl Med Mol Imaging 2007; 34: 60–67. 17. Kako S, Izutsu K, Ota Y, et al. FDG-PET in T-cell and NK-cell neoplasms. Ann Oncol 2007; 18:1685–1690. 18. Salaun PY, Gastinne T, Bodet-Milin C, et al. Analysis of 18F-FDG PET diffuse bone marrow uptake and splenic uptake in staging of Hodgkin’s lymphoma: a reflection of disease infiltration or just inflammation? Eur J Nucl Med Mol Imaging 2009; 36:1813–1821. 19. Cheson BD, Horning SJ, Coiffier B, et al. Report of an international workshop to standardize response criteria for non-Hodgkin’s lymphomas. NCI Sponsored International Working Group. J Clin Oncol 1999; 17: 1244–1253. 20. Cheson BD, Pfistner B, Juweid ME, et al. Revised response criteria for malignant lymphoma. J Clin Oncol 2007; 25:579–586. 21. Juweid ME, Stroobants S, Hoekstra OS, et al. Use of positron emission tomography for response assessment of lymphoma: consensus of the Imaging Subcommittee of International Harmonization Project in Lymphoma. J Clin Oncol 2007; 25:571–578. 22. Sugawara Y, Fisher SJ, Zasadny KR, et al. Preclinical and clinical studies of bone marrow uptake of fluorine-1-fluorodeoxyglucose with or without granulocyte colony-stimulating factor during chemotherapy. J Clin Oncol 1998; 16:173–180.
ASSESSING MARROW RESPONSE WITH FDG-PET
23. Jacene HA, Ishimori T, Engles JM, et al. Effects of pegfilgrastim on normal biodistribution of 18F-FDG: preclinical and clinical studies. J Nucl Med 2006; 47:950–956. 24. Blodgett TM, Ames JT, Torok FS, et al. Diffuse bone marrow uptake on whole-body F-18 fluorodeoxyglucose positron emission tomography in a patient taking recombinant erythropoietin. Clin Nucl Med 2004; 29: 161–163. 25. Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009; 45:228–247. 26. Wahl RL, Jacene H, Lodge MA. From RECIST to PERCIST: evolving Considerations for PET response criteria in solid tumors. J Nucl Med 2009; 50(Suppl 1). 122S–50S. 27. Freudenberg LS, Antoch G, Schu¨tt P, et al. FDG-PET/CT in re-staging of patients with lymphoma. Eur J Nucl Med Mol Imaging 2004; 31: 325–329. 28. Naumann R, Vaic A, Beuthien-Baumann B, et al. Prognostic value of positron emission tomography in the evaluation of post-treatment residual mass in patients with Hodgkin’s disease and non-Hodgkin’s lymphoma. Br J Haematol 2001; 115:793–800. 29. Lin C, Itti E, Haioun C, et al. Early 18F-FDG PET for prediction of prognosis in patients with diffuse large b-cell lymphoma: SUV-based assessment versus visual analysis. J Nucl Med 2007; 48:1626–1632. 30. Murata Y, Kubota K, Yukihiro M, et al. Correlations between 18F-FDG uptake by bone marrow and hematological parameters: measurements by PET/CT. Nucl Med Biol 2006; 33:999–1004. 31. Nahmias C, Hanna WT, Wahl LM, et al. Time course of early response to chemotherapy in non-small cell lung cancer patients with 18F-FDG PET/ CT. J Nucl Med 2007; 48:744–751. 32. Kazama T, Swanston N, Podoloff DA, et al. Effect of colony-stimulating factor and conventional- or high-dose chemotherapy on FDG uptake in bone marrow. Eur J Nucl Med Mol Imaging 2005; 32:1406–1411. 33. Lin EC. FDG PET/CT flip flop phenomenon in treated lymphoma of bone. Clin Nucl Med 2006; 31:803–805. 34. Shammas A, Lim R, Charron M. Pediatric FDG PET/CT: physiologic uptake, normal variants, and benign conditions. Radiographics 2009; 29: 1467–1486. 35. Mahmood S, Llano M. Paget disease of the humerus mimicking metastatic disease in a patient with metastatic malignant mesothelioma on whole body F-18 FDG PET/CT. Clin Nucl Med 2008; 33:510–512. 36. Bybel B. F-18 FDG PET/CT following dental extraction in a patient with head and neck cancer. Clin Nucl Med 2007; 32:741–742. 37. Shin D, Shon O, Han D, et al. The clinical efficacy of 18F-FDG-PET/CT in benign and malignant musculoskeletal tumors. Ann Nucl Med 2008; 22: 603–609. 38. Aoki J, Endo K, Watanabe H, et al. FDG-PET for evaluating musculoskeletal tumors. J Orthop Sci 2003; 8:435–441. 39. Feldman F, Heertum RV, Manos C. 18FDG PET scanning of benign and malignant musculoskeletal lesions. Skeletal Radiol 2003; 32:201–208. 40. Mester U, Even-Sapir E. Increased 18F-fluorodeoxyglucose uptake in benign, nonphysiologic lesions found on whole-body positron emission tomography/computed tomography (PET/CT): accumulated data from four years of experience with PET/CT. Semin Nucl Med 2007; 37: 206–222. 41. Dimitrakopoulou-Strauss A, Strauss LG, Heichel T, et al. The role of quantitative 18F-FDG PET studies for the differentiation of malignant and benign bone lesions. J Nucl Med 2002; 43:510–518. 42. Aoki J, Watanabe H, Shinozaki T, et al. FDG PET of primary benign and malignant bone tumors: standardized uptake value in 52 lesions. Radiology 2001; 219:774–777. 43. Grigolon MV, Delbeke D. F-18FDG uptake in a bone infarct: a case report. Clin Nucl Med 2001; 26:613–614. 44. Rosen R, Fayad L, Wahl RL. Increased 18F-FDG uptake in degenerative disease of the spine: characterization with 18F-FDG PET/CT. J Nucl Med 2006; 47:1274–1280. 45. A predictive model for aggressive non-Hodgkin’s lymphoma. The International Non-Hodgkin’s Lymphoma Prognostic Factors Project. N Engl J Med 1993; 329:987–994. 46. Schein P, Chabner B, Canellos G, et al. Non-Hodgkin’s lymphoma: patterns of relapse from complete remission after combination chemotherapy. Cancer 1975; 35:354–357. 47. Wirth A, Corry J, Laidlaw C, et al. Salvage radiotherapy for Hodgkin’s disease following chemotherapy failure. Int J Radiation Oncology Biol Phys 1997; 39:599–607.
1185