Introduction Fluorine 18 Fluorodeoxyglucose (FDG) Positron emission tomography-computed tomography (PET-CT) scan was highly sensitive modality for evaluating the staging or response of malignant lymphoma. The role of PET to detect early relapse in complete remission was still uncertain. We evaluated the clinical implication of surveillance PET-CT scan in lymphoma patients according to histologic subtype of lymphoma.

Patients and methods A total 1009 patients of 1553 treated lymphoma patients, who achieved complete response following primary treatment, were retrospectively reviewed. Among these patients, we analyzed 202 patients who showed positive results of PET-CT and performed tissue biopsy of FDG uptake sites.

Results Total 202 surveillance PET-CT positive patients, 120 (59.4%) patients were confirmed lymphoma relapse, 82 (40.6%) patients were histologically negative. Histologically negative results of 82 patients as follows: most (59.7%) patients were inflammation, 10 (2.1%) patients were solid tumor and 3 patients were tuberculosis. According to histologic subtype, PPV of Hodgkin's lymphoma was 25.0% (4/16), diffuse large B-cell lymphoma patients was 51.0% (47/92), mantle cell lymphoma (MCL) was 86.6% (13/15) and extranodal NK/T cell lymphoma (ENKL) was 85.7% (12/14) (Table 1). There was significant difference between PPV of aggressive B-cell non-Hodgkin lymphoma (NHL) except MCL and aggressive T-cell NHL (51.0% vs 78.5%, P=0.040). Indolent NHL was related with high PPV compared to aggressive B-cell NHL (70.2% vs 51.0%, P=0.032). Six (12.7%) patients of 47 indolent lymphoma were transformed to DLBCL at relapse. Four (8.5%) patients of 47 DLBCL patients showed histologic change to indolent lymphoma. There was no difference of predictive value according to stage, the number of extranodal involvement, lactic dehydrogenase and international prognostic index. PPV of FDG uptake in lymph node was 58.3% (63/108) and other sites except lymph node were 60.6% (57/94). Biopsy of stomach was 53.3% (8/15), small or large bowel was 55.5% (7/12). True positive patients of core biopsy was 61 (65.5%) of 93 patients, operation or excisional biopsy was 59 (67.0%) of 88 patients and aspiration cytology was 4 (19.0%) of 21 patients.

Conclusion PPV of surveillance PET-CT in aggressive B-cell lymphoma was lower than indolent lymphoma or aggressive T-cell lymphoma. Surveillance PET-CT had clinical limitations according to histologic subtype of lymphoma. Histologic confirm by re-biospy should be recommended to exclude the false positive in aggressive B-cell lymphoma and distinguish the histologic transformation in indolent lymphoma.

Table 1.

Positive predictive value according to lymphoma subtype

Pathology
Lymphoma subtype positive negative Positive predictive value 
DLBCL (n=92) 47 45 51.0 
MZBCL (n=26) 17 65.3 
Follicular lymphoma (n=20) 16 80.0 
Mantle cell lymphoma (n=15) 13 86.6 
Burkitt lymphoma (n=3) 33.3 
Plasmablastic lymphoma (n=1) 100 
Lymphoplasmacytic lymphoma (n=1) 
ENKL (n=14) 12 85.7 
PTCL (n=11) 63.6 
Mycosis fungoides (n=3) 66.6 
Hodgkin's lymphoma (n=16) 12 25.0 
Total (n=202) 120 82 59.4 
Pathology
Lymphoma subtype positive negative Positive predictive value 
DLBCL (n=92) 47 45 51.0 
MZBCL (n=26) 17 65.3 
Follicular lymphoma (n=20) 16 80.0 
Mantle cell lymphoma (n=15) 13 86.6 
Burkitt lymphoma (n=3) 33.3 
Plasmablastic lymphoma (n=1) 100 
Lymphoplasmacytic lymphoma (n=1) 
ENKL (n=14) 12 85.7 
PTCL (n=11) 63.6 
Mycosis fungoides (n=3) 66.6 
Hodgkin's lymphoma (n=16) 12 25.0 
Total (n=202) 120 82 59.4 

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution