Annals of Hematology

, Volume 83, Issue 5, pp 276–278

18F-fluorodeoxyglucose positron emission tomography (18F-FDG-PET) does not visualize follicular lymphoma of the duodenum

Authors

  • Martha Hoffmann
    • Department of Nuclear MedicineUniversity of Vienna
  • Andreas Chott
    • Department of Clinical PathologyUniversity of Vienna
  • Andreas Püspök
    • Department of Internal Medicine IV, Division of GastroenterologyUniversity of Vienna
  • Ulrich Jäger
    • Department of Internal Medicine I, Division of OncologyUniversity of Vienna
  • Kurt Kletter
    • Department of Nuclear MedicineUniversity of Vienna
    • Department of Internal Medicine I, Division of OncologyUniversity of Vienna
Original Article

DOI: 10.1007/s00277-003-0827-6

Cite this article as:
Hoffmann, M., Chott, A., Püspök, A. et al. Ann Hematol (2004) 83: 276. doi:10.1007/s00277-003-0827-6

Abstract

<p>18F-fluorodeoxyglucose positron emission tomography (18F-FDG-PET) has been used as a potential tool for imaging nodal follicular lymphoma (FL) and extranodal spread. As primary intestinal FL is increasingly being recognized, we have performed a study to investigate the usefulness of 18F-FDG-PET for staging of extranodal FL within the gastrointestinal tract. In eight patients with a diagnosis of FL localized in the duodenum (six cases in stage I and one each in stages II and IV, respectively) whole body 18F-FDG-PET scans were performed. Seven patients with duodenal FL were rated WHO grade 1 and one had FL grade 3, while both patients with secondary spread had FL WHO grade 1. All patients were imaged before initiation of therapy. None of the patients with primary duodenal FL showed pathologically elevated 18F-FDG uptake. 18F-FDG-PET findings were not influenced by histological grade or proliferative activity of FL. These findings suggest that 18F-FDG-PET is not useful for clinical assessment of primary duodenal FL.

Keywords

18F-FDG-PETFollicular lymphomaDuodenum

Introduction

Follicular lymphoma (FL) is one of the most common lymphoma entities, constituting about 22% of adult lymphoma cases worldwide, and up to 35–40% in the United States [1, 2]. It is characteristically composed of lymphoid cells originating in the follicle center, i.e., centrocytes and centroblasts, and displays a distinct molecular feature, the t(14;18) translocation involving rearrangement of the BCL2 gene in the large majority of cases. Upon presentation, most patients have disseminated disease affecting peripheral lymph nodes and bone marrow, while only about one-third have limited stage disease. The varying content of centroblasts and centrocytes has led to definition of three grades (grades 1–3) of FL corresponding to an increasing number of blasts per high-power field as recognized in the recent WHO classification of tumors of hematopoietic and lymphoid tissues [1]. While FL grade 1 and 2 is characterized by a relatively indolent clinical course, FL grade 3 is a more aggressive disease. A further subdivision of FL grade 3 into two distinct subtypes has recently been proposed [3].

Currently, the occurrence of FL within the gastrointestinal (GI) tract is increasingly being recognized. While data on primary GI FL are still scarce, a clustering of FL in the duodenum has been noted by various investigators [4, 5, 6]. In terms of histology, these tumors display features compatible with FL grade 1 in about 50% of cases, and may be restricted to the mucosa and submucosa without evidence of distant spread for a prolonged period of time. Accordingly, the clinical course of such localized lymphomas is thought to be highly favorable. In some patients, however, more than one segment of the GI tract may be involved by FL [4, 5, 6], necessitating a full GI work-up along with computed tomography (CT) scan of the thorax and abdomen and a bone marrow biopsy in individuals diagnosed with FL of the duodenum in order to rule out more widespread disease.

Recent years have seen the introduction of 18F-fluorodeoxyglucose positron emission tomography (18F-FDG-PET) into routine clinical practice for staging and follow-up of patients with malignant lymphoma. It is currently considered a standard imaging method in patients with Hodgkin lymphomas and individuals suffering from aggressive lymphomas, while its role in imaging indolent types of lymphoma is less clearly defined [7]. In a study on the application of 18F-FDG-PET in lymphomas of various indolent histologies, Jerusalem and co-workers [8] have also included patients with FL. In this series, the authors found a high diagnostic yield for 18F-FDG-PET especially in patients with FL, while no diagnostic benefit was seen in small lymphocytic lymphoma. In addition, it was suggested that 18F-FDG-PET might be especially useful for imaging of FL involvement at extranodal sites. In keeping with these data, Elstrom et al. [9] have reported positive 18F-FDG-PET scans in 98% of 41 patients with FL. These findings have prompted us to investigate the use of 18F-FDG-PET in patients diagnosed with FL of the duodenum, and to compare the results of nuclear imaging with endoscopy, endosonography, and CT as well as histologic assessment.

Patients and methods

A total of eight patients with a diagnosis of FL localized in the duodenum were referred to our PET unit (for patient characteristics, see Table 1). Five patients were male, while three were female, and the patients’ age was between 31 and 79 years. For staging work-up, all patients had undergone ear, nose, and throat inspection, sonography of cervical lymph nodes, CT scan of the thorax and abdomen, endosonography of the upper GI tract, gastroduodenoscopy with multiple biopsies, colonoscopy including assessment of the terminal ileum with multiple biopsies, and a bone marrow biopsy. Whole body 18F-FDG-PET scans were performed on a GE Advance PET scanner (GE Medical Systems, Waukesha, Wis., USA). PET emission scanning was started 40 min following the intravenous bolus injection of about 380 MBq of 18F-FDG and was performed from the head to the pelvic floor. Emission scans of 5 min acquisition time and subsequent transmission scans of 4 min (~75,000,000 counts) per bed position were obtained. Transmission data were smoothed by a one-dimensional filter in axial direction and segmented, and attenuation correction factors were calculated by forward projection. The images were reconstructed iteratively. Imaging with 18F-FDG was performed before initiation of therapy in all patients, and 18F-FDG-PET results were compared with results of conventional imaging as stated above and histologic findings.
Table 1

Characteristics of patients with duodenal FL. MCP mitoxantrone, chlorambucil, and prednisone, R-CHOP rituximab, cyclophosphamide, vincristine, and prednisone

Sex/age

Stage

Grading

Ki-67/MIB-1

Treatment

M/40

I

1

30%

MCP, radiation

M/39

I

1

10%

Radiation

F/79

I

1

15%

Radiation

F/53

I

1

10%

Radiation

M/31

IVa

1

20%

None

F/52

II

1

5%

Radiation

M/35

I

3

80%

Radiation, R-CHOP

M/70

I

1

15%

Radiation

aDue to discrete bone marrow involvement

Results

All eight patients with duodenal FL had the diagnosis established by histologic work-up of samples obtained during endoscopy performed due to unspecific epigastric complaints. In terms of grading, seven patients had features compatible with FL grade 1, while one patient was rated as FL grade 3. While the small amount of tissue present in the bioptic samples did not reliably allow for counting of centroblasts in ten neoplastic follicles as stated in the WHO classification [1], the assumption of correct histological grading appears reasonable in view of the fact that all patients rated as grade 1 had <5 centroblasts per high-power field. A relatively low proliferative activity characteristic for FL as judged by reactivity with the Ki-67 antibody (MIB-1) was seen in all patients with FL grade 1, while the patient with FL grade 3 had a MIB-1 reactivity of about 80% (see Table 1). Macroscopically, all patients had multiple small polypoid lesions clustering around the papilla, while one also had a small duodenal ulcer measuring 1.5 cm in diameter. Endosonography, however, revealed involvement of the whole duodenal circumference in this patient, which was nevertheless restricted to the mucosa and submucosa. A similar pattern with lymphomatous spread restricted to mucosa and submucosa was seen on endosonography in four patients, while endosonography was normal in one patient. These six patients were rated as being in stage I, while involvement of local lymph nodes compatible with stage II was seen in another patient. In the patients whose lymphoma was visible on endosonography, the lesions measured between 15×20 mm to a maximum of 20×25 mm. No further manifestations of FL indicating generalized disease beyond local lymph nodes were found by means of extensive staging in these seven individuals. In another patient with lymphoma restricted to mucosa and submucosa on endosonography, also involvement of the bone marrow was found, corresponding to stage IV disease.

The 18F-FDG-PET scans of the patients with duodenal FL showed no pathologically elevated focal FDG uptake in the abdomen or in distant sites, irrespective of histological grade or proliferative activity. In one patient 18F-FDG-PET showed a diffusely elevated uptake along the whole colon, while the documented lesion involving the whole duodenal circumference did not display an increased 18F-FDG accumulation. Multiple biopsies obtained during endoscopy nevertheless did not reveal lymphomatous involvement in the colon and the terminal ileum. In addition, also no signs of inflammatory changes, which might have resulted in unspecific elevated tracer accumulation, were detected. The intensity of 18F-FDG uptake in this patient was in the range reported as a result of smooth muscle contractions. Accordingly, the results of 18F-FDG-PET scanning were rated negative in all patients.

Discussion

Primary intestinal FL is a rare condition, and has only recently been recognized as having distinct clinical and pathological features [4, 5, 6]. According to the data generated so far, there appears to be a striking clustering of FL in the duodenum, especially around the papilla, and the clinical course of such patients is thought to be relatively indolent. Yet, in some patients more than one segment of the GI tract may be affected, as may be extraintestinal organs and lymph nodes in widespread disease. In such cases, however, it may be impossible to distinguish between primary or secondary involvement of the GI tract. As treatment and long-term prognosis are based on the clinical stage upon presentation, staging procedures should include ear, nose, and throat inspection, sonography of cervical lymph nodes, CT scan of the thorax and abdomen, endosonography of the upper GI tract, gastroduodenoscopy with multiple biopsies, colonoscopy including assessment of the terminal ileum with multiple biopsies, and a bone marrow biopsy. Thus, a method which is able to visualize the whole body and reliably detect distant lymphoma involvement would definitely facilitate the staging work-up and thus influence the subsequent therapeutic decisions.

18F-FDG-PET is a well-established method in the staging and follow-up of various types of (predominantly aggressive) lymphomas. Based on two recent studies [8, 9], which found a high clinical impact of 18F-FDG-PET in FL, we have evaluated the ability of 18F-FDG-PET to visualize duodenal FL. Naturally, our series includes only a small number of patients, as primary intestinal FL appears to be a very rare disease. As opposed to the promising data obtained by Jerusalem et al. [8] and Elstrom et al. [9] in nodal FL (albeit with a slightly different imaging protocol), our results were negative, as none of the patients displayed 18F-FDG uptake within the lymphoma. In addition, one of these lymphomas also could not be seen on endosonographic imaging. In this patient, however, one might speculate that the lymphoma was simply too small to be determined by 18F-FDG-PET. In the remaining patients, however, the tumor mass should in theory have been sufficiently above the detection threshold according to the diffuse involvement of mucosa and submucosa in all patients with additional lymph nodes in another case. As judged by endosonography, the lesions measured between 15×20 mm to up to 20×25 mm. The superficial involvement, i.e., restriction to mucosa and submucosa, however, still might have led to underestimation of 18F-FDG uptake in patients with stage I disease. These data suggest that histologic assessment of biopsy samples remains the diagnostic gold standard for the time being. Judging from our small series, neither histologic grading nor proliferative activity offer a potential explanation for the negative findings in duodenal FL, while FL of nodal origin can apparently be visualized on 18F-FDG-PET according to recent data [8, 9]. Of interest is the fact that a similar situation as in duodenal FL is seen with a more common type of extranodal lymphoma, i.e., extranodal marginal zone B-cell lymphoma of the mucosa-associated lymphoid tissue (MALT) usually located in the GI tract [10]. Both diseases share common features, i.e., superficial involvement of GI organs in a high percentage, a restriction to mucosal sites for a prolonged time, and an indolent clinical course. Taken together, our findings demonstrate that 18F-FDG-PET does not visualize duodenal FL as opposed to nodal FL. Such patients should still undergo extensive conventional staging work-up.

Copyright information

© Springer-Verlag 2003