Original Research
Nuclear Medicine and Molecular Imaging
August 2010

Characterization of T-Cell Lymphomas by FDG PET/CT

Abstract

OBJECTIVE. The purpose of this article is to describe the utility of FDG PET/CT in documenting sites of disease in patients with T-cell lymphomas, to quantify the degree of FDG avidity in the various subtypes of this heterogeneous group of disorders, and to highlight the pattern of imaging findings associated with specific disease subtypes.
MATERIALS AND METHODS. A retrospective review of patients with T-cell lymphomas who underwent PET/CT examination for initial disease staging or at disease relapse over a 5-year period was undertaken by correlation between a patient database and a PACS radiology information system. Disease subtypes were grouped according to World Health Organization categorization of mature natural killer cell–T-cell neoplasms. Sites of disease involvement were documented according to cutaneous or extranodal, nodal, and visceral locations. The maximum standardized uptake value (SUV) was recorded for each patient.
RESULTS. One hundred thirty-five patients with T-cell lymphoma were included, and sites of disease were documented by use of FDG PET/CT in 122 (90%) patients. Of those 122 patients, 55 (45%) had cutaneous involvement, 95 (78%) had FDG-avid lymphadenopathy, and 54 (44%) had FDG-avid extranodal disease other than cutaneous involvement. A significant difference in maximum SUV was observed in cases of mycosis fungoides and mycosis fungoides with large cell transformation (SUV, 11.3 vs 3.8; p < 0.05).
CONCLUSION. We found high rates of FDG positivity in T-cell lymphoma. Given the propensity for disease involvement outside the normal scan range of diagnostic CT, we recommend that patients with T-cell lymphoma be scanned from vertex to feet by use of PET/CT.

Introduction

T-cell lymphomas comprise a heterogeneous group of uncommon lymphomas, accounting for 12% of all cases of non-Hodgkin lymphoma [1], or approximately 5,000 new cases per year diagnosed in the United States [2]. Considerable geographic variability has been reported in the distribution of peripheral T-cell lymphoma, varying from 1.5% of cases of non-Hodgkin lymphoma in Vancouver to 18.3% of cases in Hong Kong [3]. Despite their relative rarity, the T-cell lymphomas are subdivided by the World Health Organization into more than 20 recognizable clinicopathologic entities; this classification system was updated in 2008. As a group, T-cell lymphomas have a poorer prognosis than do B-cell lymphomas [4].
T-cell lymphomas show a spectrum of clinical behavior, from chronic indolent to aggressive variants; extranodal involvement is frequently seen. A description of the radiologic features associated with specific disease subtypes has been made [5]; however, the characteristics of T-cell lymphomas on FDG PET/CT are not well established. Several case reports [610] and a recent literature review [11] suggest FDG avidity; however, comprehensive data are lacking. There has been a belief that FDG PET/CT is less useful in the assessment of T-cell lymphoma than in other types of lymphoma, on the basis of reports of low rates of FDG positivity [1214]. However, those studies were limited by the small number of patients. For instance, a large series of 172 patients with various histologic subtypes of lymphoma included only five patients with peripheral T-cell lymphoma; in two of those five patients, FDG-positive disease was detected, yielding a sensitivity of 40% [12]. This lack of sufficient information prompted our retrospective analysis, with the aim of highlighting typical disease patterns in the various disease subsets, organized according to the World Health Organization's classification of mature T-cell–natural killer (NK) cell neoplasms. Given the retrospective nature of our study, a true assessment of the sensitivity and specificity of FDG PET/CT in T-cell lymphoma cannot be made; however, we highlight the high rate of FDG positivity among patients with T-cell lymphoma referred for PET/CT.

Materials and Methods

This retrospective analysis was performed under a waiver from the institutional review board; patient consent was not required. Using a database of patients with T-cell lymphoma and correlation with the PACS, we retrospectively identified patients who underwent FDG PET/CT during their clinical course at our institution. Correlation with the electronic medical record system was made. A single FDG PET/CT for each patient was eligible for inclusion if it was performed at the time of initial diagnosis or suspected relapse. We excluded patients who had scans obtained midtreatment, or immediately after completion of treatment, and patients who underwent PET/CT at an outside institution. The following data were recorded: age, sex, histologic subtype of T-cell lymphoma, study indication, PET positivity, sites of disease, and maximum standardized uptake value (SUV). PET positivity was defined as a site of abnormal FDG uptake in tissue histologically proven or clinically or radiographically suspected to represent lymphomatous involvement. The sites of disease were characterized as either nodal or extranodal, and specific sites of nodal and extranodal disease were recorded. A single maximum SUV representing the maximum intensity voxel within a volumetric region of interest was recorded for each patient, measured at the site of highest FDG uptake in that scan. The cases were grouped according to specific World Health Organization categories.
Fig. 1 Distribution of maximum standardized uptake value ranges in T-cell lymphoma grouped according to World Health Organization's classification of natural killer cell–T-cell neoplasms. Significant difference in maximum standardized uptake values was observed between total number of mycosis fungoides (MF) and Sézary syndrome cases and MF with large cell transformation. Boxes contain data points from the 25th to the 75th percentile. Diamonds indicate means, and vertical lines within boxes indicate medians. Whiskers (lines to left and right side of boxes) extend from 10% to 90% of entire data distribution. AITL = angioimmunoblastic T-cell lymphoma, ALCL = anaplastic large cell lymphoma, ATLL = adult T-cell leukemia lymphoma, EATCL = enteropathy-associated T-cell lymphoma, NK-TCL = natural killer cell–T-cell lymphoma, PTCLU = peripheral T-cell lymphoma, unclassified, SPTCL = subcutaneous panniculitis-like T-cell lymphoma.

FDG PET/CT

All patients underwent FDG PET/CT on one of four PET/CT systems without specific preference (Biograph16, Siemens Healthcare Molecular Imaging; or Discovery LS, Discovery ST, or Discovery STE, all from GE Healthcare). Images were acquired after IV injection of 15 mCi (555 MBq) FDG and a 60- to 90-minute uptake period. Blood glucose level was less than 200 mg/dL before injection. Oral contrast medium (meglumine diatrizoate, Gastrografin, Bristol-Myers Squibb; 2.5% solution, 1,000 mL) was administered. After CT (120 kV, 40–80 mA), PET emission images were obtained at 3 minutes per bed position. Patients with a diagnosis of cutaneous T-cell lymphoma were imaged from the vertex to the feet; otherwise, the image range extended from midskull to upper thighs. PET, CT, and fusion images were reviewed on a workstation integrated with a PACS (Centricity AW Suite version 2, GE Healthcare) that allowed multiplanar reformatting of images. Slight differences in resolution and quantification can be observed between the different PET/CT systems; however, we have observed ≤ 10% differences in maximum SUV provided that all other parameters, such as uptake time and image reconstruction parameters, are standardized.

Statistical Analysis

Maximum SUVs across subtypes were compared using a one-way analysis of variance, and the pairwise p values were adjusted for multiple comparisons using the bootstrap [15]. These data are also summarized in the form a box plot (Fig. 1). A p value of less than 0.05 was considered statistically significant.

Results

The study population comprised a database of 190 patients with T-cell lymphoma. Eight patients did not undergo PET/CT, 14 underwent PET/CT at an outside institution, and 33 underwent FDG PET/CT during or after completion of treatment; all of these patients were excluded from the final analysis. Our total eligible population comprised 135 patients (79 men and 56 women), with a mean age of 57 years (range, 21–90 years). The ratio of men to women was 1.4:1. One hundred twenty-two (90%) of the 135 patients showed abnormal FDG uptake in at least one site. Of those 122 patients, 55 (45%) had abnormal cutaneous or subcutaneous uptake, 95 (78%) had FDG-avid lymphadenopathy, and 54 (44%) had FDG-avid extranodal disease other than cutaneous involvement. A total of 87 cases (71%) showed extranodal involvement combining cutaneous and visceral disease. The summary of the patient distribution and imaging findings is shown in Table 1.
TABLE 1: Summary of PET/CT Findings in All Cases of T-Cell Lymphoma
Sex, No. of PatientsPET Scan Positive, No. of Patients/Total (%)Type of Disease, No. of Patients/TotalDisease Outside Field of View,a No. of PatientsMaximum Standardized Uptake Value, Mean (Range)
Type of LymphomaNo. of PatientsMaleFemaleCutaneousNodalVisceral
PTCLU34241033/34 (97)8/3329/3317/33712.3 (2.8-42.3)
Systemic ALCL1611515/16 (94)4/1515/158/15322.9 (5.3-51.0)
AITL1871114/18 (78)0/1414/147/14012.6 (4.8-29.2)
Mycosis fungoides126610/12 (83)9/104/100/1053.8 (1.4-8.9)
Sézary syndrome8358/8 (100)5/88/81/825.0 (3.4-10.2)
Transformed mycosis fungoides115610/11 (91)9/108/104/10711.3 (2.3-25.0)
Extranodal NK-TCL1211110/12 (83)6/106/107/10510.8 (4.9-23.3)
SPTCL9729/9 (100)9/91/91/965.7 (1.5-13.1)
ATLL9279/9 (100)3/98/96/9415.5 (2.3-34.8)
Cutaneous ALCL4312/4 (50)2/21/21/208.05 (4.2-11.9)
EATCL
2
0
2
2/2 (100)
0/2
1/2
2/2
0
19.9 (14.7-25.0)
Total
135
79
56
122/135 (90)
55/122
95/122
54/122
39

Note—AITL = angioimmunoblastic T-cell lymphoma, ALCL = anaplastic large cell lymphoma, ATLL = adult T-cell leukemia lymphoma, EATCL = enteropathy-associated T-cell lymphoma, NK-TCL = extranodal natural killer cell—T-cell lymphoma, PTCLU = peripheral T-cell lymphoma, unclassified, SPTCL = subcutaneous panniculitis-like T-cell lymphoma.
a
Refers to sites of disease outside of the field of view of a CT scan of the neck, chest, abdomen, and pelvis.

FDG Avidity

Sites of active disease were clearly detected in all histologic subtypes. The fraction of PET-positive patients varied within each histologic subgroup, as well as the intensity of FDG uptake (Fig. 1). The highest SUVs were recorded in patients with aggressive variants, as specified in Figure 1. Of note, there was a significant difference in the mean maximum SUV in patients with mycosis fungoides with transformation (SUV, 11.3) compared with patients with mycosis fungoides without transformation (SUV, 3.82; p < 0.05). In addition, the SUVs were significantly higher in systemic anaplastic large cell lymphoma than in mycosis fungoides, angioimmunoblastic T-cell lymphoma, NK cell–T-cell lymphoma, subcutaneous panniculitis-like T-cell lymphoma, and peripheral T-cell lymphoma, unclassified (p < 0.02). There was no significant difference in maximum SUV between anaplastic large cell lymphoma and human T-cell lymphotropic virus (HTLV)–associated T-cell lymphoma or enteropathy-associated T-cell lymphoma.

CT Correlation

Thirty-nine patients (29%) had sites of disease beyond the range of a diagnostic neck, chest, abdomen, and pelvis CT scan. These sites included cutaneous, subcutaneous, and muscular masses of the scalp, upper and lower extremities, and lymphadenopathy in epitrochlear and popliteal regions.

Imaging Features of Nodal T-Cell Lymphomas

Peripheral T-cell lymphoma, unspecified—The designation of peripheral T-cell lymphoma, unspecified, includes all lymphomas of mature postthymic T-cell origin, without specific clinicopathologic features that permit further classification. Sixty percent of patients have stage IV disease at the time of diagnosis [16]. We encountered 34 patients with this disease, 33 of whom had FDG-avid disease. Eight (24%) of 33 patients had cutaneous or subcutaneous lesions. Twenty-nine patients (88%) showed FDG-avid lymphadenopathy, 17 (51%) had visceral disease, and, in four patients, second primary malignancies were detected incidentally, involving the breast, lung, colon, and epiglottis. The range of maximum SUVs for all sites was 2.8–42.3 (mean, 12.3), indicating the heterogeneous and predominantly FDG-avid nature of this disease.
Anaplastic large cell lymphoma—Primary systemic anaplastic large cell lymphomas may be further classified into two subtypes, anaplastic lymphoma kinase positive and negative. Expression of the anaplastic lymphoma kinase protein confers a more favorable prognosis than the other peripheral T-cell lymphomas [16]. Our population included 16 patients with systemic anaplastic T-cell lymphoma. In four cases (25%), we noticed cutaneous or subcutaneous focal FDG uptake. One patient had multiple subcutaneous nodules along the lymphovascular drainage pathway from the lower extremity (Fig. 2). Increased FDG uptake in lymph nodes was seen in 15 patients (94%), with SUVs between 3.3 and 37.0. Eight (50%) patients had visceral sites of disease, specifically involving lung (n = 3) and muscle, bone, liver, tonsil, and parotid gland (n = 1 each). There was no obvious association between anaplastic lymphoma kinase–positive (n = 4) or anaplastic lymphoma kinase–negative (n = 12) state and the FDG avidity of tumors.
Angioimmunoblastic T-cell lymphoma— Angioimmunoblastic T-cell lymphoma is a distinct subtype of peripheral T-cell lymphoma that manifests clinically with lymphadenopathy, hepatosplenomegaly, skin rash, and constitutional symptoms, with evidence of dysregulation of the immune system. Many patients show hypergammaglobulinemia, Coombs-positive hemolytic anemia, and evidence of immunodeficiency [16]. Our study population included 18 patients with angioimmunoblastic T-cell lymphoma at initial staging or relapse, 14 of whom had FDG-avid disease detected on PET. None of these 14 patients had hypermetabolic skin lesions, but all showed FDG-avid lymphadenopathy (average maximum SUV, 12.3; range, 4.1–29.2). Seven patients had visceral involvement (in some cases at multiple sites), including the spleen (n = 4), bones (n = 2), lung (n = 2), multifocal sites in the gastrointestinal tract (n = 1), liver (n = 1), or nasopharynx (n = 1).
Fig. 2 54-year-old man with anaplastic large cell lymphoma. Three-dimensional maximum intensity projection from FDG PET/CT image of lower extremities shows multiple hypermetabolic foci correlating with soft-tissue nodules along lymphovascular drainage pathway of right leg (standardized uptake value range, 3.5–14.8).
Fig. 3 25-year-old man with extranodal natural killer cell–T-cell lymphoma, nasal type. Three-dimensional maximum-intensity-projection, axial PET, axial PET/CT fused, and axial CT images from FDG PET/CT show FDG-avid mass in nasal region. Innumerable thick cutaneous hypermetabolic plaques (standardized uptake value range, 1.5–12.7) (arrow and arrowhead) were evident throughout rest of body.
Fig. 4 55-year-old woman with mycosis fungoides. Plaque as small as 1-cm lesion can be seen (standardized uptake value, 1.4).
Fig. 5 44-year-old woman with mycosis fungoides tum uptake value, 8.9).
Fig. 6 79-year-old woman with ulcerated exophytic scalp tumor of mycosis fungoides. Three-dimensional maximum-intensity-projection, sagittal PET, sagittal PET/CT fused, and axial CT images are shown (standardized uptake value, 5.6).

Imaging Features of Extranodal Visceral T-Cell Lymphomas

Extranodal NK cell–T-cell lymphoma, nasal type—Extranodal NK cell–T-cell lymphoma, nasal type, is the most common peripheral T-cell lymphoma in Asian populations and has been previously called both angiocentric lymphoma and lethal midline granuloma [5]. Epstein-Barr virus infection may play a role in tumorigenesis. The tumor often involves the nasal cavity or paranasal sinuses, and tumor necrosis is a common feature [17]. Identical tumors can develop in an extranasal location (hence, the designation “nasal type”), including the skin, gastrointestinal tract, testis, kidneys, and respiratory system.
We reviewed PET scans for 12 patients with extranodal NK cell–T-cell lymphoma, 10 of whom had true-positive FDG-avid disease. Only one patient had a primary nasal tumor (Fig. 3). Six patients had predominantly cutaneous disease, varying from a solitary cutaneous lesion to innumerable cutaneous and subcutaneous masses. Visceral involvement was seen in seven cases, involving lung (n = 3), bone marrow (n = 2), spleen (n = 2), and intestine, liver, and pericardium (n = 1). Nodal sites of disease were also seen in six patients.
Enteropathy-associated T-cell lymphoma— Enteropathy-associated T-cell lymphoma is a rare lymphoma that typically occurs in elderly patients with a history of celiac disease or dermatitis herpetiformis. The diagnosis of celiac disease may not have been established before the diagnosis of lymphoma; however, pathologic specimens usually have evidence of villous atrophy in the adjacent nonlymphomatous bowel [4].
Fig. 7 28-year-old man with Sézary syndrome. Whole-body coronal reformat of FDG PET/CT shows diffusely increased cutaneous FDG uptake (maximum standardized uptake value, 3.2) and right popliteal hypermetabolic adenopathy. This pattern of uptake appears similar to that seen in nonattenuation-corrected emission scans.
Fig. 8 63-year-old man with transformed mycosis fungoides. Three-dimensional maximum-intensity-projection, axial PET, axial fused PET/CT, and axial CT images show markedly hypermetabolic heterogeneous skin thickening (standardized uptake value range, 7.0–14.3) and FDG-avid axillary adenopathy.
Two patients with this rare diagnosis underwent FDG PET/CT. For one patient, the PET/CT scan showed an FDG-avid ileal mass and mesenteric adenopathy. The other patient had already undergone ileal resection, but the postsurgical staging PET/CT scan revealed two FDG-avid small intestinal foci, which had not been seen optimally on either a contrast-enhanced CT or small-bowel follow-up study.

Imaging Features of Extranodal Cutaneous T-Cell Lymphomas

Mycosis fungoides—Mycosis fungoides is a cutaneous T-cell lymphoma attributed to the clonal expansion of epidermotropic CD4+ helper T cells that exists in three clinical phases: patches, plaques, and tumors [18]. There were 12 patients with mycosis fungoides in our study; PET/CT detected local cutaneous disease in nine cases, comprising one or more FDG-avid plaques or tumors (Table 1). Patches (n = 2) were only appreciated clinically and were not amenable to radiologic detection. Mycosis fungoides tumors are known to be visible on CT scans [5, 19]; however, we were able to detect abnormal FDG uptake associated with both plaque and tumor disease in nine of 10 cases on PET/CT, when particular attention was devoted to the skin. Cutaneous disease in mycosis fungoides had a spectrum of radiologic appearances that varied from areas of cutaneous soft-tissue thickening to frank exophytic soft-tissue masses (Figs. 4, 5, 6). We found hypermetabolic adenopathy in only four of the 12 patients with plaque or tumor stage disease. No patient showed visceral involvement.
Fig. 9 43-year-old man with subcutaneous panniculitis-like T-cell lymphoma. Three-dimensional maximum-intensity-projection image, axial PET, axial PET/CT fused, and axial CT image from FDG PET/CT show three intensely FDG-avid subcutaneous infiltrative lesions with ill-defined borders (arrows) represent sites of lymphoma (standardized uptake value range, 5.0–13.1).
Sézary syndrome—Sézary syndrome is a leukemic variant of mycosis fungoides, in which characteristic Sézary cells are found in the peripheral blood. These cells are the result of a clonal proliferation of malignant T cells; they show a loss of normal cell surface antigens and characteristic hyperconvoluted cerebriform nuclei. Clinically, diffuse erythroderma is seen [18]. Our study population included eight patients with Sézary syndrome. Only two of six clinically erythrodermic patients showed a pattern of diffusely increased cutaneous FDG uptake. This pathologic finding simulates the artifact of diffuse superficial FDG uptake seen in nonattenuation-correction PET emission images (Fig. 7). Three of the eight patients exhibited focal FDG-avid skin lesions (SUV range, 3.4–10.2). FDG-avid lymphadenopathy was seen in all eight patients (SUV range, 3.5–6.2). Low-grade FDG uptake in lymph nodes is of uncertain clinical significance in Sézary syndrome because benign dermatopathic lymphadenitis commonly occurs. This term refers to the reactive enlargement of lymph nodes draining areas of disruption of cutaneous integrity, thus explaining the false-positive FDG uptake on PET. Of our eight patients with Sézary syndrome with FDG-avid lymph nodes, only two had biopsyproven malignant involvement, one had probable malignant lymphadenopathy on the basis of clinical assessment, and five had likely false-positive FDG uptake secondary to dermatopathic lymphadenopathy, on the basis of clinical assessment. One patient had hypermetabolic infiltration of skeletal muscle; otherwise, there was no visceral disease.
Fig. 10 49-year-old man with subcutaneous panniculitis-like T-cell lymphoma. Coronal and axial PET images, axial PET/CT fused, and axial CT image from FDG PET/CT show hypermetabolic soft-tissue infiltrates subcutis diffusely, in pattern considered characteristic of this disease.
Mycosis fungoides with large cell transformation—Mycosis fungoides may undergo transformation to a CD30+ anaplastic large cell lymphoma with aggressive biologic behavior in up to 39% of patients [20]. We encountered 11 cases of mycosis fungoides with large cell transformation, and 10 had true-positive FDG PET/CT scans. Nine of these patients had markedly FDG-avid cutaneous lesions (Fig. 8). The average maximum SUV in this group of patients (SUV, 11.3) was significantly higher than that in the group of patients with mycosis fungoides without transformation (SUV, 3.8) and that for patients with mycosis fungoides and Sézary syndrome (SUV, 5.0; p < 0.05). Eight of the 10 patients also had hypermetabolic lymphadenopathy. Visceral involvement was observed frequently, with parotid gland involvement in two patients, diffuse splenic involvement in two patients, and hypermetabolic infiltration of the gracilis muscle in the lower extremity in one patient.
Subcutaneous panniculitis-like T-cell lymphoma—In subcutaneous panniculitis-like T-cell lymphoma, which is a rare distinct primary lymphoma of the subcutaneous tissues, multifocal subcutaneous infiltrates of malignant T lymphocytes are associated with benign histiocytic inflammatory cells. A characteristic histopathologic feature is the rimming of individual fat cells by neoplastic lymphocytes [21]. Clinically, numerous indurated poorly circumscribed firm violaceous cutaneous plaques and nodules develop. Subcutaneous panniculitis-like T-cell lymphoma shows a spectrum of clinical behavior; a more aggressive hemophagocytic syndrome is sometimes seen, manifested by coagulopathy, B symptoms, and hepatosplenomegaly, and it invariably has a fatal outcome [4].
The unique imaging findings of subcutaneous panniculitis-like T-cell lymphoma are not widely described [5, 10]. CT depicts multifocal areas of soft-tissue subcutaneous infiltration with poorly defined margins. In our experience, these lesions range in size from a few centimeters to confluent infiltration of the entire subcutis of the trunk and lower extremities (Figs. 9 and 10). Infiltration of the omental fat was also seen in one patient. The degree of soft-tissue infiltration of the adipose tissue is variable, ranging from negative Hounsfield units, indicating significant residual fat, to average soft-tissue Hounsfield unit values, indicating predominant soft-tissue component. On metabolic imaging, these lesions are invariably FDG avid. All of our nine cases showed multifocal hypermetabolic subcutaneous infiltrative lesions (Table 1). Pulmonary infiltrates were observed in one case, featuring multiple ground-glass nodules between 0.5 and 1.5 cm diameter. Only one of nine patients had a solitary hypermetabolic inguinal lymph node. Subcutaneous panniculitis-like T-cell lymphoma lesions could potentially be misinterpreted as sites of subcutaneous injection if the anterior abdominal wall or gluteal region is involved. The differential diagnosis also includes inflammatory disorders of the subcutaneous tissues, such as erythema nodosum or erythema induratum, as well as inflammatory or infectious panniculitis [5].
Primary cutaneous anaplastic large cell lymphoma—Primary cutaneous anaplastic large cell lymphoma belongs to a spectrum of disease that includes the benign entity lymphomatoid papulosis, and differentiation between these conditions can be challenging for the clinician and the pathologist. Patients usually present with nonregressing cutaneous nodules 1–2 cm in size [4]. In up to 10% of cases, spread to extracutaneous sites may occur [4]. Our study population included four cases of primary cutaneous anaplastic large cell lymphoma. Two cases showed false-positive FDG uptake: one with increased focal tracer activity in the calf (maximum SUV, 2.9) and the other with focal suspicious uptake at the site of a previously irradiated skin lesion (maximum SUV, 3.4); biopsy results were negative for lymphoma. The other two cases were true-positives: One patient showed an area of mild FDG uptake in a skin lesion in the flank (SUV, 2.2). In the other patient, PET/CT demonstrated true-positive FDG-avid disease progression to extracutaneous sites involving the lungs and mediastinal lymph nodes.

Leukemic Disseminated T-Cell Lymphoma

Adult T-cell leukemia lymphoma (HTLV associated) is a distinct peripheral T lymphocyte malignancy associated with HTLV type 1, a retrovirus endemic in the Caribbean and Japan [4]. We reviewed FDG PET/CT in nine patients at the time of initial staging or relapse, and all showed sites of FDG-avid disease (Fig. 1). One third of patients had FDG-avid cutaneous and subcutaneous lesions, and eight of nine had FDG-avid lymphadenopathy (Fig. 1). Extranodal disease was observed in six patients, involving the spleen (n = 3); bone (n = 2); and the liver, nasopharynx, nose, lung, and parotid gland (n = 1).

Discussion

Our study suggests a clear role for FDG PET/CT in the evaluation of T-cell lymphomas. In distinction to previous reports, the great majority of our patients with various forms of T-cell lymphoma showed FDG-avid disease. FDG avidity was detected throughout the range of T-cell histologies. Among the histologies reported here with eight or more cases, FDG avidity ranged from 78% in angioimmunoblastic T-cell lymphoma to 90–100% in anaplastic large cell lymphoma, subcutaneous panniculitis-like T-cell lymphoma, and adult T-cell leukemia lymphoma. PET was less predictably positive among the cutaneous anaplastic large cell lymphomas; however, only four cases were included in our study.
In this series, 71% of patients had extranodal disease; in 48% of cases, this was outside of the range covered by the routine staging examination—that is, a CT scan of the chest, abdomen, and pelvis. Even if a neck CT had been part of the routine staging process, in 30% of cases these disease sites would still have been outside the field of view. FDG PET/CT identifies more disease sites than conventional CT does because the scan range is greater. The combination of PET with CT allows greater lesion detection because foci of subtle FDG uptake in skin and subcutaneous tissues may be overlooked unless the simultaneously acquired CT images of the PET/CT are reviewed carefully with particular attention to these areas. This finding highlights the need to review CT images independently “in their own right” to define the extent of T-cell lymphoma properly. Systemic anaplastic large cell lymphoma seems to be the most FDG-avid variant of T-cell lymphoma. However, the other aggressive variants, including peripheral T-cell lymphoma, unclassified, adult T-cell leukemia lymphoma, and extranodal NK cell–T-cell lymphoma, also appear consistently FDG avid.
The SUVs in our series are comparable with those described elsewhere. For instance, Bishu et al. [22] reported a wide SUV range of 1–23 g/mL in six patients with peripheral T-cell lymphoma, unclassified. Khong et al. [23] confirmed that extranodal NK cell–T-cell lymphoma, angioimmunoblastic T-cell lymphoma, peripheral T-cell lymphoma, unclassified, and systemic anaplastic large cell lymphoma are consistently FDG avid in a series of 26 patients. Another study reported SUVs ranging from 5 to 25 in 10 patients with extranodal NK cell–T-cell lymphoma [17]. We believe that the wide range of SUVs in these disease entities reflects tumor biology and local aggressiveness, although many other factors, such as tumor stage, volume of disease, and local blood supply, may also contribute [24].
We have found FDG PET/CT to be useful in the evaluation of cutaneous disease in mycosis fungoides. In prior studies, there have been conflicting reports on the detection rate of cutaneous T-cell lymphoma on FDG PET. In two smaller series [13, 14], citing nine and three cases, respectively, PET scans were either completely negative in patients with stage Ia disease or showed a high false-negative rate [13]. However, one study reported an overall accuracy of 87% of FDG PET in a heterogeneous group of 19 patients with cutaneous lymphoma [25]. In addition, at least one report exists of true-positive mycosis fungoides plaque disease on PET/CT [26]. Of note, the studies using PET-only scanning systems appear to have a lower detection rate. We propose that the combination of PET with CT contributes to the improved detection of cutaneous sites of disease in mycosis fungoides because combined PET/CT clearly distinguishes between focal FDG uptake in skin as opposed to artifacts or uptake in nodes and subcutaneous tissues. Overall, FDG PET/CT appears useful in defining the local extent of disease and in characterization of mycosis fungoides of plaque or tumor stage type. We concur with previous investigators who found skin disease to be less well characterized in Sézary syndrome [27]; despite clinical erythrodermic state, we observed detectable FDG uptake in only one third of these patients.
In contrast to a prior study [27], we have shown a statistically significant difference in maximum SUV between mycosis fungoides and mycosis fungoides with large cell transformation. This finding is similar to our prior experience with FDG in B-cell lymphoma [28]. Although sites of cutaneous disease will be apparent clinically, FDG PET/CT offers the advantage of characterizing the metabolic activity in a lesion. We propose that biopsies be guided to the most FDG-avid site of skin disease if large cell transformation is suspected. PET/CT may also guide biopsies of FDG-avid lymph nodes because it is important to distinguish tumor involvement from nonspecific dermatopathic reactive changes [27]. False-positive metabolic activity may also be seen in mycosis fungoides as a result of lesion ulceration and superinfection. In subcutaneous panniculitis like T-cell lymphoma, the mechanism of FDG uptake is likely bifactorial; histologically, malignant T lymphocytes are surrounded by benign inflammatory cells.
Patients with cutaneous T-cell lymphomas may be at increased risk for second primary malignancies [29]. Because FDG PET/CT detected four incidental cancers in the peripheral T-cell lymphoma, unclassified, subtype in our series, we propose that this is an effective method of detecting occult solid malignancies in this patient population. Of the detected second malignancies, only the malignant lung nodule would have been readily identified on a diagnostic CT scan; the malignancies of the epiglottis, breast, and colon were not identifiable.
Our study had some limitations related to its retrospective nature and the small numbers of patients in some of the histologic subgroups. Our standard since 2004 has been to include PET/CT as part of initial staging for all patients with aggressive T-cell lymphomas. For patients with indolent or skin disease, PET/CT is performed in all patients with stage IIB (tumors) or higher, or stage IA–IB with palpable nodes or systemic symptoms. Thus, although this analysis is retrospective, the great majority of patients would have received PET/CT unless their insurance refused to authorize. Therefore, we estimate that any referral bias is minimal. The primary objective of this article was a description of imaging findings in all T-cell lymphomas, a topic that was previously not well covered in the literature. A dedicated assessment of sensitivity, specificity, and effect on patient management was beyond the scope of this project. It should be stressed that the SUVs quoted in this article primarily provide an orientation and should not be interpreted in isolation for diagnosis or classification of abnormalities. Although we strive to reproduce each FDG PET/CT under the exact same conditions, inevitably in this retrospective review there will be variations in maximum SUV related to use of different scanning systems and minor variations in uptake time; however, we estimate that these values would not vary by greater than 10%. Finally, an assessment of the utility of FDG PET/CT in monitoring the response to therapy and in prognostication was beyond the scope of this article.
In summary, we have highlighted the spectrum of imaging findings in T-cell lymphomas and have emphasized the high rate of FDG positivity in most entities, suggesting that FDG PET/CT should be part of the routine workup of these patients. In contrast to Hodgkin lymphoma and B-cell lymphoma, the imaging findings can be subtle and the disease can involve unusual sites, such as skin, muscle, or viscera, that are beyond the normal field of diagnostic CT. Therefore, at least at primary staging, patients with T-cell lymphoma should undergo FDG PET/CT extending from the vertex to the feet to identify occult sites of disease, which may potentially change disease stage and patient management.

Footnote

Address correspondence to J. Feeney ([email protected]).

References

1.
[No authors listed]. A clinical evaluation of the International Lymphoma Study Group classification of non-Hodgkin's lymphoma. The Non-Hodgkin's Lymphoma Classification Project. Blood 1997; 89:3909 –3918
2.
Armitage JO, Vose JM, Weisenburger DD. Towards understanding the peripheral T-cell lymphomas. Ann Oncol 2004; 15:1447 –1449
3.
Rüdiger T, Weisenburger DD, Anderson JR, et al. Non-Hodgkin's Lymphoma Classification Project. Peripheral T-cell lymphoma (excluding anaplastic large-cell lymphoma): results from the Non-Hodgkin's Lymphoma Classification Project. Ann Oncol 2002; 13:140–149
4.
Savage KJ. Peripheral T-cell lymphomas. Blood Rev 2007; 21:201 –216
5.
Lee HJ, Im JG, Goo JM, et al. Peripheral T-cell lymphoma: spectrum of imaging findings with clinical and pathologic features. RadioGraphics 2003; 23:7–26
6.
George CD, Wilson AG, Philpott NJ, Bevan DH. The radiological features of adult T-cell leukaemia/lymphoma. Clin Radiol 1994; 49:83 –88
7.
Shapiro M, Yun M, Junkins-Hopkins JM, et al. Assessment of tumor burden and treatment response by 18F-fluorodeoxyglucose injection and positron emission tomography in patients with cutaneous T- and B-cell lymphomas. J Am Acad Dermatol 2002; 47:623–628
8.
Watanabe N, Kato H, Murakami J, et al. FDG-PET imaging in cutaneous anaplastic large cell lymphoma. Clin Nucl Med 2006; 31:564 –565
9.
Watanabe N, Murakami J, Kameda K, et al. F-18 FDG-PET imaging in adult T-cell leukemia lymphoma. Clin Nucl Med 2008; 33:423 –425
10.
Ravizzini G, Meirelles GS, Horwitz SM, Grewal RK. F-18 FDG uptake in subcutaneous panniculitis-like T-cell lymphoma. Clin Nucl Med 2008; 33:903 –905
11.
Otero HJ, Jagannathan JP, Prevedello LM, et al. CT and PET/CT findings of T-cell lymphoma. AJR 2009; 193:349–358
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.
Valencak J, Becherer A, Der-Petrossian M, Trautinger F, Raderer M, Hoffmann M. Positron emission tomography with [18F] 2-fluoro-D-2-deoxyglucose in primary cutaneous T-cell lymphomas. Haematologica 2004; 89:115–116
14.
Kako S, Izutsu K, Ota Y, et al. FDG-PET in T-cell and NK-cell neoplasms. Ann Oncol 2007; 18:1685 –1690
15.
Westfall PH, Young SS. Resampling-based multiple testing. New York, NY: Wiley, 1993
16.
Vose JM. Peripheral T-cell Non Hodgkin's lymphoma. Hematol Oncol Clin North Am 2008; 22:997–1005
17.
Karantanis D, Subramaniam RM, Peller PJ, et al. The value of [(18)F]fluorodeoxyglucose positron emission tomography/computed tomography in extranodal natural killer/T-cell lymphoma. Clin Lymphoma Myeloma 2008; 8:94 –99
18.
Smith BD, Wilson LD. Cutaneous lymphoma. Curr Probl Cancer 2008; 32:43 –87
19.
Malloy PC, Fishman EK, Magid D. Lymphoma of bone, muscle, and skin: CT findings. AJR 1992; 159:805–809
20.
Diamandidou E, Colome-Grimmer M, Fayad L, et al. Transformation of mycosis fungoides/Sézary syndrome: clinical characteristics and prognosis. Blood 1998; 92:1150 –1159
21.
Gallardo F, Pujol RM. Subcutaneous panniculitic-like T-cell lymphoma and other primary cutaneous lymphomas with prominent subcutaneous tissue involvement. Dermatol Clin 2008; 26:529–540
22.
Bishu S, Quigley JM, Schmitz J, et al. F-18-fluoro-deoxy-glucose positron emission tomography in the assessment of peripheral T-cell lymphomas. Leuk Lymphoma 2007; 48:1531 –1538
23.
Khong PL, Pang CB, Liang R, Kwong YL, Au WY. Fluorine-18 fluorodeoxyglucose positron emission tomography in mature T-cell and natural killer cell malignancies. Ann Hematol 2008; 87:613 –621
24.
Lucignani G, Paganelli G, Bombardieri E. The use of standardized uptake values for assessing FDG uptake with PET in oncology: a clinical perspective. Nucl Med Commun 2004; 25:651–656
25.
Kumar R, Xiu Y, Zhuang HM, Alavi A. 18F-fluorodeoxyglucose-positron emission tomography in evaluation of primary cutaneous lymphoma. Br J Dermatol 2006; 155:357 –363
26.
Kuo PH, McClennan BL, Carlson K, et al. FDG-PET/CT in the evaluation of cutaneous T-cell lymphoma. Mol Imaging Biol 2008; 10:74 –81
27.
Tsai EY, Taur A, Espinosa L, et al. Staging accuracy in mycosis fungoides and Sézary syndrome using integrated positron emission tomography and computed tomography. Arch Dermatol 2006; 142:577 –584
28.
Noy A, Schöder H, Gönen M, et al. The majority of transformed lymphomas have high standardized uptake values (SUVs) on positron emission tomography (PET) scanning similar to diffuse large B-cell lymphoma (DLBCL). Ann Oncol 2009; 20:508–512
29.
Brownell I, Etzel CJ, Yang DJ, Taylor SH, Duvic M. Increased malignancy risk in the cutaneous T-cell lymphoma patient population. Clin Lymphoma Myeloma 2008; 8:100–105

Information & Authors

Information

Published In

American Journal of Roentgenology
Pages: 333 - 340
PubMed: 20651187

History

Submitted: September 22, 2009
Accepted: January 22, 2010
First published: November 23, 2012

Keywords

  1. FDG PET/CT
  2. large cell transformation
  3. mycosis fungoides
  4. standardized uptake value
  5. subcutaneous panniculitis-like T-cell lymphoma
  6. T-cell lymphoma

Authors

Affiliations

John Feeney
Department of Radiology, Nuclear Medicine Service, Memorial Sloan-Kettering Cancer Center, 1275 York Ave., Box 77, New York, NY 10065.
Steven Horwitz
Department of Medicine, Lymphoma Service, Memorial Sloan-Kettering Cancer Center, New York, NY.
Mithat Gönen
Department of Biostatistics, Memorial Sloan-Kettering Cancer Center, New York, NY.
Heiko Schöder
Department of Radiology, Nuclear Medicine Service, Memorial Sloan-Kettering Cancer Center, 1275 York Ave., Box 77, New York, NY 10065.

Metrics & Citations

Metrics

Citations

Export Citations

To download the citation to this article, select your reference manager software.

Articles citing this article

Media

Figures

Other

Tables

Share

Share

Copy the content Link

Share on social media