Abstract 1402

Poster Board I-424

Introduction:

Anatomic imaging using contrast-enhanced computed tomography (CT) is essential for management of lymphomas. Functional imaging using 18FDG-PET (PET) improves detection of certain lymphomas, specifically, diffuse large B-cell lymphoma (DLBCL) and follicular lymphoma (FL). Currently, PET imaging is performed with co-registration of low-dose non-contrast enhanced CT images used for anatomic correlation and attenuation correction of PET images (PET/CT). It has been suggested that the low-dose non-enhanced CT cannot substitute for diagnostic contrast-enhanced CT imaging since the arterial and venous phases of contrast enhancement improve detection of lesions. Given the differential sensitivity for detection of specific lymphomas by PET imaging, we hypothesized that FDG could substitute for intravenous contrast in imaging of certain lymphomas, and that PET/CT or PET imaging could potentially obviate the need for contrast-enhanced CT. To test this hypothesis, we performed an independent and blinded radiology review of these imaging studies in patients (pts) with DLBCL, FL, small lymphocytic lymphoma/chronic lymphocytic leukemia (CLL/SLL) or marginal zone lymphoma (MZL) who had contrast-enhanced CT, PET, PET/CT, and low-dose non-enhanced CT available for retrospective review.

Patients and Methods:

Pts with a diagnosis of DLBCL, FL, CLL/SLL, or MZL with PET/CT and contrast-enhanced CT studies performed at the Hospital of the University of Pennsylvania within 6 weeks of each other without intervening therapy were studied. Pts with clinically suspected progression of lymphoma between studies were excluded. Radiologists, blinded to clinical information or other imaging results, separately interpreted image sets of low-dose non-enhanced CT, PET, fusion PET/CT, and contrast-enhanced CT studies. The presence or absence of disease at 44 nodal and 48 (female) or 49 (male) extranodal sites was recorded for each site for each imaging modality. Concordant findings across imaging modalities were defined as positive for involvement by lymphoma; discordant findings were reconciled using all available clinical and radiologic information with follow-up for progression or regression of abnormality, or by biopsy.

Results:

Between May 2006 and January 2008, 55 pts with either DLBCL (n=31), FL (n=13), CLL/SLL (n=5), or MZL (n=6) had complete images sets available for review. All patients had at least 18 months of clinical follow-up after imaging. A total of 282 sites met criteria for involvement by lymphoma. The rates of detection for specific lymphomas by each imaging modality are shown below:

Sensitivity by Histology and Imaging Modality
.Contrast CTLow-dose CTPETPET/CT
DLBCL 63% (56/89*) 55% (49/89) 74% (66/89) 91% (81/89) 
    nodal 71% (47/66) 53% (35/66) 77% (51/66) 91% (60/66) 
    extranodal 39% (9/23) 61% (14/23) 65% (15/23) 91% (21/23) 
FL 78% (76/98) 52% (51/98) 65% (64/98) 90% (88/98) 
    nodal 80% (70/88) 57% (50/88) 73% (64/88) 91% (80/88) 
    extranodal 60% (6/10) 10% (1/10) 0% (0/10) 80% (8/10) 
MZL 86% (19/22) 50% (11/22) 36% (8/22) 64% (14/22) 
    nodal 90% (18/20) 55% (11/20) 35% (7/20) 65% (13/20) 
    extranodal 50% (1/2) 0% (0/2) 50% (1/2) 50% (1/2) 
CLL/SLL 89% (65/73) 64% (47/73) 22% (16/73) 55% (40/73) 
    nodal 95% (60/63) 71% (45/63) 17% (11/63) 54% (34/63) 
    extranodal 50% (5/10) 20% (2/10) 50% (5/10) 60% (6/10) 
Sensitivity by Histology and Imaging Modality
.Contrast CTLow-dose CTPETPET/CT
DLBCL 63% (56/89*) 55% (49/89) 74% (66/89) 91% (81/89) 
    nodal 71% (47/66) 53% (35/66) 77% (51/66) 91% (60/66) 
    extranodal 39% (9/23) 61% (14/23) 65% (15/23) 91% (21/23) 
FL 78% (76/98) 52% (51/98) 65% (64/98) 90% (88/98) 
    nodal 80% (70/88) 57% (50/88) 73% (64/88) 91% (80/88) 
    extranodal 60% (6/10) 10% (1/10) 0% (0/10) 80% (8/10) 
MZL 86% (19/22) 50% (11/22) 36% (8/22) 64% (14/22) 
    nodal 90% (18/20) 55% (11/20) 35% (7/20) 65% (13/20) 
    extranodal 50% (1/2) 0% (0/2) 50% (1/2) 50% (1/2) 
CLL/SLL 89% (65/73) 64% (47/73) 22% (16/73) 55% (40/73) 
    nodal 95% (60/63) 71% (45/63) 17% (11/63) 54% (34/63) 
    extranodal 50% (5/10) 20% (2/10) 50% (5/10) 60% (6/10) 
*

= sites detected / sites involved by lymphoma

Conclusions:

Our results suggest that combined PET/CT imaging is more sensitive than contrast-enhanced CT imaging for detection of DLBCL and at least as sensitive as contrast-enhanced CT imaging for detection of FL. In comparison, contrast-enhanced CT imaging appears superior to PET/CT imaging for CLL/SLL; while further studies are needed to confirm superiority of contrast-enhanced CT imaging in MZL. The routine use of both contrast enhanced CT and PET/CT modalities for staging of lymphoma may be unnecessary, potentially increasing both the cost of medical care and radiation exposure. Additional studies are needed to determine which imaging modality is optimal for each type of lymphoma.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution