The proliferation potential of tumor cells is indicated by the proliferation of Ki-67, a nuclear antigen expressed by dividing cells. A high Ki-67 proliferation index (MIB-1-labeling index) is associated with poor prognosis in non-Hodgkin lymphoma, T-cell lymphoma, and localized extranodal natural killer (NK)/T-cell lymphoma, nasal type (ENKL). The maximum standardized uptake value (SUVmax) of the biopsy site during 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) reflects tumor aggressiveness in non-Hodgkin lymphoma cases. In ENKLs, a 100% positive rate of FDG-PET has been previously reported. Here, we evaluated the biopsy sites of patients with untreated ENKL to clarify the correlation between SUVmax at the biopsy site and the proliferation potential of tumor cells. Between 1998 and 2011, 22 patients were newly diagnosed with ENKL at Yokohama City University Hospital and Kanagawa Cancer Center. In 17 cases, the tumors were staged using PET or PET/computed tomography (CT), and biopsies were simultaneously conducted. Variations in SUVmax among institutions and the underestimations derived from small tumors, which are limitations of PET carried out in multicenter studies, were corrected by a phantom study by using a NEMA IEC Body Phantom SetTM. The difference in the accuracy of SUVmax between the two institutions was less than 2%. We studied 15 extranodal biopsy specimens from 13 patients with untreated ENKL, for whom complete data were available. The biopsy specimens were reviewed by two hematopathologists (S.S. and K.T.), according to the World Health Organization (WHO) classification. The study comprised 10 men and 3 women, with a median age of 52 years (range: 18–82 years). We found that the SUVmax was not correlated with the MIB-1- labeling index (r = − 0.33, P = 0.22). In conclusion, this study indicates that the SUVmax during FDG-PET does not predict the proliferation potential of ENKLs.