Positron emission tomography/computed tomography in the diagnosis,staging,and prognostic evaluation of natural killer/T-cell lymphoma

Abstract

Natural killer/T-cell lymphoma (NKTL) is a rare subtype of non-Hodgkin’s lymphoma that is associated with Epstein–Barr virus infection. The clinicopathological features of NKTL are unique among lymphomas. NKTL is an aggressive disease with a poor prognosis in the absence of effective treatment. Accurate diagnosis and staging are essential to ensure an appropriate treatment strategy and accurate prognosis of NKTL. ¹⁸F-Fluorodexoyglucose-positron emission tomography/computed tomography (¹⁸F-FDG PET/CT) is a valuable technique in the diagnosis, staging, and prognostic evaluation of various types of malignant tumors, including NKTL. PET/CT imaging studies of patients with NKTL have shown that NKTL is ¹⁸F-FDG-avid and that PET/CT is superior to conventional methods in detecting cutaneous and extracutaneous lesions. We herein review recent PET/CT studies that have provided considerable insight into the diagnosis, staging, prognostic evaluation, and treatment effectiveness in patients with NKTL.

Keywords

Natural killer/T-cell lymphoma non-Hodgkin’s lymphoma diagnosis 18F-fluorodexoyglucose-positron emission tomography/computed tomography prognosis staging

Introduction

Natural killer/T-cell lymphoma (NKTL) is an uncommon subtype of non-Hodgkin’s lymphoma that is usually associated with Epstein–Barr virus (EBV) infection.¹ Patients with NKTL are mainly located at East Asia, Southeast Asia, and Latin America; conversely, NKTL is relatively unusual in Europe and North America.^2–4 Clinically, NKTL is divided into two types: extranodal nasal type NKTL (ENKTL) and extranasal NKTL. ENKTL is characterized by frequent necrosis, angiocentric growth, and a cytotoxic phenotype and usually affects the nasal cavity, nasopharynx, and upper aerodigestive tract.⁵ Extranasal NKTL frequently involves multiple areas including the skin, gastrointestinal tract, testis, and soft tissue.⁶ ENKTL is typically associated with a poorer treatment response and prognosis than other types of lymphomas.⁷ Chemotherapy in combination with radiotherapy is usually needed to achieve superior outcomes.⁸ The cumulative probability of 5-year survival ranges from 37.9% to 49.5%.⁹ Accurate diagnosis and staging are essential in the treatment strategy and prognosis of NKTL. However, neither an optimal treatment nor useful prognostic factors have been determined.

¹⁸F-Fluorodexoyglucose-positron emission tomography/computed tomography (¹⁸F-FDG PET/CT) has been widely used in the staging, prognosis, and treatment outcomes of various types of malignant tumors, including Hodgkin’s lymphoma and various types non-Hodgkin’s lymphoma.^10–13 Because of the poorer treatment response and prognosis of ENKTL, accurate diagnosis and staging of NKTL is urgently needed. We herein review recent PET/CT studies that have provided considerable insight into the diagnosis, staging, prognostic evaluation, and treatment effectiveness in patients with NKTL.

Diagnostic utility of PET/CT in patients with NKTL

PET/CT is a valuable tool for assessing extranodal involvement in patients with NKTL.^13–18 In one study, the mean ¹⁸F-FDG uptake measured by the maximum standardized uptake value (SUV_max) was found to be significantly higher in NKTL.¹⁷ To calculate the SUV_max, the region of interest was drawn along the margin of the lesion. The SUV_max was obtained using the following formula: SUV_max = maximum activity in the region of interest (MBq/g)/[injected dose (MBq)/body weight (g)]).

In 2010, Wu et al.¹⁷ retrospectively analyzed 15 patients with NKTL and assessed the role of 18F-FDG PET/CT in staging. ¹⁸F-FDG PET/CT detected nasal or extranasal lymphoma lesions in at least one site in all 15 patients. Of 11 patients with ENKTL, high levels of ¹⁸F-FDG uptake were observed in the nasal cavities in 8 patients and in the nasopharynx in 2 patients, and extranasal lesions were found in 7 patients with nasal type NKTL. In four patients with ENKTL, no obvious ¹⁸F-FDG uptake was seen in either the nasal cavity or nasopharynx; however, multiple extranasal lesions were identified.¹⁷ These results suggest that NKTL exhibits high ¹⁸F-FDG uptake and that PET/CT is a useful tool in the staging of this disease.

In 2011, Fujiwara et al.¹⁵ compared the utility of PET/CT with conventional methods (including CT, biopsy, and bone marrow examination) in the staging of ENKTL. Nineteen untreated patients with ENKTL were analyzed, and 116 lesions were detected by conventional methods and PET/CT. In total, 108 lesions (93%) were discovered by PET/CT and only 80 lesions were detected using conventional methods.¹⁵ The number of extranodal lesions detected by conventional methods and PET/CT was 89; 84 (94%) and 51 (61%) lesions were positive as detected by PET/CT and conventional methods, respectively.¹⁵ In a similar study, Liu et al.¹⁶ evaluated the utility of ¹⁸F-FDG PET/CT in the diagnosis of cutaneous ENKTL. In total, 39 patients with newly diagnosed ENKTL were included. PET/CT and conventional methods detected 139 lesions, among which 50 were cutaneous and 89 were extracutaneous-positive lesions.¹⁶ ¹⁸F-FDG PET/CT detected 48 cutaneous and 88 extracutaneous lesions, while conventional methods detected only 34 cutaneous lesions and 61 extracutaneous lesions that were positive for malignancy.¹⁶

A similar study was performed in a larger group of patients with ENKTL.¹⁸ In total, 1300 anatomic lesions were assessed with an ¹⁸F-FDG PET/CT scan and with conventional methods.¹⁸ Only 59 nodal and 71 extranodal anatomic lesions were truly positive for malignancy. PET/CT detected 58 nodal and 69 extranodal anatomic lesions that were malignant, whereas conventional methods detected only 44 nodal and 61 extranodal anatomic lesions that were malignant.¹⁸ Thus, PET/CT exhibited a significantly better sensitivity and specificity than conventional methods for the detection of malignant lesions. Moreover, PET/CT findings altered the original staging category for 12 patients (21.2%) and affected treatment planning in 23 patients (44.2%).¹⁸ These studies demonstrated that PET/CT is superior to conventional methods in detecting cutaneous and extracutaneous lesions. Zhou et al.¹⁹ assessed the role of FDG-PET/CT in bone marrow involvement in patients with ENKTL and found that the sensitivity and specificity of FDG-PET/CT for identifying bone marrow involvement was 100% and 86%, respectively. They suggested that FDG-PET/CT may be used as a complementary tool in patients with bone marrow involvement not detected by bone marrow biopsy. Thus, PET/CT is a useful tool for staging and treatment planning in patients with NKTL.

PET/CT also has limitations, including false-positive results. Liu et al.¹⁶ reported 2 false-positive lesions of 429 lesions in 39 patients with ENKTL detected by PET/CT. One lesion with high FDG uptake was found in the right abdominal wall skin, but the biopsy revealed herpes zoster. The other lesion was found on the left lower limb, and the biopsy confirmed inflammatory cell infiltration; this lesion disappeared after 1 month.¹⁶ Cheson et al.²⁰ reported that the false-positive rate of interim PET was 87% and that the positive predictive value was only 32% in patients with diffuse large B-cell lymphoma. They suggested that PET/CT scans should not be performed for at least 3 weeks, preferably 6 to 8 weeks, after completion of therapy because this false-positive result may persist for up to 2 weeks after chemotherapy alone or for 2 to 3 months after radiation therapy or chemoradiotherapy.²⁰ In another study of patients with cutaneous NKTL, the positive rate of FDG-PET was only 50%.²¹ The authors suggested that there was a limited role of FDG-PET in detection of cutaneous lesions of T/NK cell neoplasms because of the partial volume averaging effect.²¹

Staging and restaging roles of PET/CT in patients with ENKTL

Recent studies have suggested that PET/CT may be more advantageous than a routine work-up in ENKTL staging. ¹⁸F-FDG PET/CT may be more effective than conventional methods for detecting nodal and extranodal malignant lesions and for correctly identifying primary sites missed by conventional methods.^15–18,22 Moreover, the original staging may be modified and the treatment strategies may be influenced by PET/CT data.^15–18 However, because limited numbers of patients were included in previous studies, it is difficult to conclude that FDG PET/CT should serve as the standard staging method.^15–18

Wu et al.¹⁷ found that 18F-FDG PET/CT detected more lesions than did conventional methods (biopsies, regional CT or magnetic resonance imaging, B-ultrasound, chest radiography, and medical examination). 18F-FDG PET/CT imaging of patients with ENKTL changed the stage of disease in six patients: four were upstaged and two were downstaged.¹⁷ Fujiwara et al.¹⁵ compared the utility of PET/CT and that of conventional methods in the staging of ENKTL. The results showed that PET/CT findings altered the stage and treatment strategy in two cases (11%). MacDonald et al.²³ reported that PET/CT improved target volume delineation and aided the staging of and radiotherapy planning in the treatment of ENKTL. Moon et al.¹⁸ reported that PET/CT findings altered the original staging category in 12 patients with ENKTL (21.2%) and affected treatment planning in 23 patients (44.2%). Moreover, PET/CT exhibited significantly better sensitivity than conventional methods for the detection of malignant lesions.¹⁸ Liu et al.¹⁶ found that 18F-FDG PET/CT provided more accurate staging than did conventional methods in patients with cutaneous NKTL. In total, 39 patients with cutaneous NKTL were assessed using an ¹⁸F-FDG PET/CT scan and conventional methods, and the results showed that ¹⁸F-FDG PET/CT staging was consistent with the final stage determination (biopsy and clinical follow-up) in 94.9% (37/39) of patients, whereas staging by conventional methods was correct in the final stage determination in 74.4% (29/39) of patients.¹⁶ These results indicate that ¹⁸F-FDG PET/CT is a valuable modality for staging and treatment planning in patients with ENKTL.

Evaluation of NKTL treatment responses with PET/CT

Recent studies have confirmed the beneficial role of PET/CT in monitoring treatment responses in patients with non-Hodgkin’s lymphoma, including NKTL, which is FDG-avid. Bai et al.²⁴ showed that the SUV_max predicted the responses to primary treatment in 81 patients with ENKL. A higher SUV_max (31.3) predicted treatment failure and was associated with bulky disease, local invasion, and a high Korean Prognostic Index score.

The Deauville score (DS) can also be used to assess treatment responses. The DS is both accurate and reproducible, affording good interobserver agreement.²⁵ Khong et al.²⁶ used mid-treatment 18F-FDG PET/CT to prospectively evaluate the responses of patients with ENKTL to the standardized SMILE chemotherapy regimen (prednisolone, methotrexate, ifosfamide, L-asparaginase, and etoposide). ¹⁸F-FDG PET/CT was useful for assessing responses during the early to middle phases of treatment. The study findings suggested that the 5-point DS allows continuous evaluation of treatment responses.²⁶ Kim et al.¹³ found that the post-treatment DS and EBV-DNA status allowed patients with NKTL to be stratified by the risk associated with treatment. All patients were classified into a low- or high-risk group. The low-risk group comprised EBV-DNA-negative patients with a post-treatment DS of 1 or 2, and the high-risk group comprised EBV-DNA-positive patients with a DS of 1 or 2 and EBV-DNA-negative patients with a DS of 3 or 4 at the end of treatment. Continuous treatment was recommended for the high-risk patients, while the low-risk patients underwent follow-up only.¹³ A similar study also provided evidence by showing that sophisticated patient selection using PET/CT scanning and whole-blood EBV-DNA might provide additional information in the treatment of ENKTL.²⁷

Use of PET/CT to evaluate the prognosis of patients with NKTL

Many factors affect the prognosis of patients with NKTL,^26–33 including advanced-stage disease (stage III or IV), an unfavorable international prognostic index score, a poor Korean Prognostic Index score,³³ bone or skin invasion, an elevated level of circulating EBV-DNA, an elevated lactate dehydrogenase level, a higher body mass index at the time of diagnosis, and the presence of EBV-positive cells in the bone marrow.^34,35 However, reliable prognostic factors for NKTL remain controversial. New prognostic models continue to be developed; these include the prognostic index of natural killer lymphoma (PINK) and the PINK plus EBV-DNA models,³⁶ featuring PET/CT data including the SUV_max,^24,28,29 and models such as the whole-body metabolic tumor volume (WBMTV) and the whole-body level of total lesional glycolysis (WBTLG).³⁰ (The WBMTV was determined from attenuation-corrected PET data using software according to the following procedure. First, a rectangular parallelepiped-shaped volume of interest fully encasing all involved lesions in the axial, coronal, and sagittal FDG PET/CT images was drawn. The boundaries of voxels with an SUV intensity exceeding 3.0 were then produced automatically. Second, normal organs including the brain, heart, stomach, liver, intestines, kidney, ureter, bladder, and brown adipose tissue were manually subtracted from the product of the previous step. Third, false-positive lesions, such as inflammation, infection, or other benign FDG-avid lesions based on histopathological reports or other imaging modalities, were subtracted. Finally, WBTLG was calculated as the summation of individual metabolic tumor volume (MTV) multiplied by its SUV_mean of every ENTKL lesion.) Several studies have shown that the SUV and textural features may also predict the prognosis of NKTL.^31,33 However, the SUV_max reflects only the most obvious metabolic activity of a tumor in a region of interest and therefore depicts only the maximum metabolic rates of small regions of certain tumors, not the total tumor metabolism. Thus, use of the SUV_max alone to predict the prognosis may be quantitatively misleading because FDG uptake may vary according to the individual patient with NKTL, the reference background, and the PET/CT system employed. New models for assessment of the prognosis of NKTL continue to be developed. MTV and total lesional glycolysis (TLG) are easily calculated by new software and reflect the whole metabolic tumor burden. TLG is an ideal metabolic parameter that combines the SUV_mean and MTV to combine assessments of tumor volume and metabolism.^30,31 However, several limitations exist: some studies include few patients with ENKTL; measurement of MTV is not reliable, and reproducibility is low, especially for multiple disseminated lesions; and a standard MTV threshold has not been established.^30,31,37

As early as 2008, Suh et al.²⁸ evaluated whether the pretreatment 18F-FDG uptake was a predictor of survival in patients with ENKTL. Multivariate analysis revealed that only the SUV_max of the primary site independently predicted disease-specific survival.²⁸ However, their study had certain limitations, including a short follow-up time, a retrospective design, and a small number of patients. Bai et al.²⁴ also found that the pretreatment SUV_max is predictive of the prognosis in patients with newly diagnosed ENKTL. Jiang et al.²⁹ prospectively investigated the prognostic utility of pretreatment ¹⁸F-FDG uptake and interim and post-therapy PET/CT data in 33 patients with ENKTL. The multivariate analysis revealed that the SUV_max of the primary tumor and post-therapy PET/CT data were prognostic in terms of both progression-free survival (PFS) and overall survival (OS). However, the interim PET/CT data were not significantly predictive of survival.²⁹

Kim et al.³⁰ were the first to explore whether the SUV_max, WBMTV, and WBTLG of pretreatment FDG PET/CT images were predictive of the prognosis of patients with ENKTL. The WBMTV and WBTLG were higher in patients with than without progressive disease. The WBMTV and WBTLG of patients who died were higher than those of survivors. The multivariate analysis revealed that an SUV_max of >8.1, WBMTV of >14.4 cm³, and WBTLG of >52.7 cm³ were significantly prognostic in terms of PFS. Of these factors, a WBMTV of >14.4 cm³ was the best prognostic factor in terms of OS.³⁰

Song et al.³¹ explored whether the MTV as determined by PET/CT (a measure of the lymphoma burden) was prognostic in 80 patients with I_E/II_E stage NKTL. The 3-year PFS and OS of patients with high MTVs were lower than those of patients with low MTVs when a value of 35.2 cm³ was used as the MTV cut-off. The multivariate analysis revealed that a good disease status and upfront radiotherapy were independently predictive of both PFS and OS.³¹ The addition of radiotherapy to chemotherapy was suggested to benefit patients with high tumor burdens.³¹ Liang et al.³⁸ suggested that the SUV_max reflects the highest metabolic rate at only a single lymphoma site and was thus not representative of the metabolic rates at all lymphoma sites. Although the WBMTV and WBTLG may better reflect the metabolic status of all lymphoma sites, this is still disputed. Therefore, the cited authors introduced three new models using the SUVs of different sites. The WB1SUV_max model gives the sum of the whole-body SUV_max values of 11 nodal and 10 extranodal lesions. The WB2SUV_max model gives the sum of the whole-body SUV_max values of 3 nodal (neck, axillary, and inguinal and spleen) and 10 extranodal lesions. The WB3SUV_max model gives the sum of the whole-body SUV_max values of 3 nodal (superior diaphragm, inferior diaphragm, and spleen) and 10 extranodal lesions. Receiver operating characteristic curves revealed that the optimal cut-off values for the WB1SUV_max, WB2SUV_max, and WB3SUV_max models were 15.8 (sensitivity, 92%; specificity, 67%; area under the curve [AUC], 0.811; p < 0.001), 12.7 (sensitivity, 96%; specificity, 57%; AUC, 0.785; p < 0.001), and 15.8 (sensitivity, 88%; specificity, 70%; AUC, 0.793; p < 0.001), respectively. The WB1SUV_max, WB2SUV_max, and WB3SUV_max models were all significantly better than the SUV_max model. The WB3SUV_max model was selected for further prognostic investigation. The multivariate analysis showed that the WB3SUV_max value was an independent prognostic factor.³⁸

Some authors have found that the DS may be prognostic of NKTL. Khong et al.²⁶ used mid-treatment DS cut-offs of 1 to 3 (compared with 4 to 5) when performing receiver operating characteristic analysis of OS. The DS was the only predictor of both OS and PFS and was thus more powerful than other measures.²⁶ Kim et al.¹³ used the following 5-point DS scale to assess 102 patients with NKTL after completion of planned treatment: 1 (absence of lesion uptake), 2 (low lesion uptake or uptake equivalent to that of the mediastinum), 3 (high lesion uptake by the mediastinum but low or equivalent uptake by the liver), 4 (moderately high liver uptake), and 5 (markedly high uptake in the liver with development of new lesions). A post-treatment DS of 3 or 4 and EBV-DNA positivity were independently prognostic. The results suggested that the DS is easily calculated, reproducible, and associated with good interobserver agreement when used to assess the prognosis of NKTL.¹³ Lim et al.²⁷ assessed the prognostic utility of the pretransplant DS (using the above-mentioned 5-point scale) in 27 patients with ENKTL who underwent autologous stem cell transplantation followed by PET/CT combined with assessment of whole-body EBV-DNA positivity. A DS cut-off of 1 to 2 was used to evaluate survival. Two groups of patients were created according to their pretransplantation DS and EBV-DNA data: a favorable risk group (DS of 1 or 2 and EBV-DNA negativity) and an unfavorable risk group (DS of 3‒5 and EBV-DNA negativity, DS of 1 or 2 and EBV-DNA positivity, or DS 3‒5 and EBV-DNA positivity). The estimated 3-year OS and PFS rates were 67% and 44%, respectively, in the first group, but only 39% and 15%, respectively, in the second group.²⁷ The multivariate analysis revealed that the DS and EBV-DNA positivity were the only independent prognostic factors.²⁷ The prospective study by Jiang et al.³² yielded similar results. The cited authors explored the prognostic utility of 3 different models in 60 patients with ENKTL: the International Harmonization Project (IHP) model, the Deauville 5-point scale, and SUV-based assessment. The IHP model was not predictive of either PFS or OS. It was suggested that FDG is not a specific tracer and can be absorbed by inflammatory cells and yield false-positive results on interim PET/CT. All FDG uptake ranges (very low uptake, significant residual uptake but less than pretreatment uptake, no change in uptake, and uptake progression) were associated with high false-positive rates. Univariate analyses revealed that the interim PET/CT outcomes based on the DS and change in SUV_max were predictors of survival. The multivariate analysis showed that the DS was prognostic of PFS and OS and that the change in SUV_max was the only significant predictor of OS.³² In an another study, the authors assessed the ability to predict prognosis using the IHP criteria, Deauville 5-point scale, and change in SUV_max in 59 patients with ENKTL. They found that the Deauville 5-point scale was more valuable for predicting prognosis than was the IHP or change in SUV_max in patients with ENKTL.³⁹ However, Khong et al.²⁶ found that the change in SUV_max was not significantly prognostic of PFS or OS. Differences in the therapies employed in the two studies may partly explain the conflicting results.

Ko et al. found that the following textural features were associated with PFS: contrast, dissimilarity, high-intensity short-zone emphasis, high-intensity zone emphasis (HIZE), low-intensity short-zone emphasis (LISZE), busyness, coarseness, black/white symmetry, and run-length variability.³¹ The univariate analysis showed that the extent of cancer, contrast, dissimilarity, HIZE, LISZE, coarseness, and black/white symmetry were predictive of PFS. The multivariate analysis revealed that dissimilarity and LISZE were the principal predictors of PFS. However, neither SUV_max, WBMTV, nor WBTLG predicted the prognosis of NKTL.³¹ The differences among these studies may be attributed to differences in the sample size, type of NKTLs evaluated, SUV threshold, or SUV_max cut-off. All of the studies reviewed above were limited by small sample sizes and single-center designs; prospective multicenter studies are needed.

A recent meta-analysis of the prognostic value of PET/CT suggested that multiple parameters of PET/CT were more valuable than a single parameter for assessment of the prognosis of patients with ENKTL.³⁷

Conclusions

¹⁸F-FDG PET/CT may play pivotal roles in the diagnosis, staging, restaging, and evaluation of treatment outcomes in patients with NKTL. The combination of many parameters with the SUV_max of PET/CT may be more valuable for prognostic evaluation in these patients.

Footnotes

Author contributions

JJD and SHZ designed the review and wrote the manuscript. YLC reviewed some of the literature. KZ reviewed some of the literature and revised the manuscript.

Declaration of conflicting interest

The authors declare that there is no conflict of interest.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Cai

Luo

Zhang

et al . New prognostic model for extranodal natural killer/T cell lymphoma, nasal type. Ann Hematol 2014; 93: 1541–1549. doi:10.1007/s00277-014-2089-x.

William

Armitage

JO.

International analysis of the frequency and outcomes of NK/T-cell lymphomas.

Best Pract Res Clin Haematol 2013; 26: 23–32. doi:10.1016/j.beha.2013.04.003.

Abouyabis

Shenoy

Lechowicz

et al . Incidence and outcomes of the peripheral T-cell lymphoma subtypes in the United States. Leuk Lymphoma 2008; 49: 2099–2107. doi:10.1080/10428190802455867.

Gill

Liang

Tse

Extranodal natural-killer/t-cell lymphoma, nasal type.

Adv Hematol 2010; 2010: 627401. doi:10.1155/2010/627401.

Tse

Kwong

YL.

Practical management of natural killer/T-cell lymphoma.

Curr Opin Oncol 2012; 24: 480–486. doi:10.1097/CCO.0b013e3283556142.

Chan

Sin

Wong

et al.

Nonnasal lymphoma expressing the natural killer cell marker CD56: a clinicopathologic study of 49 cases of an uncommon aggressive neoplasm. Blood 1997; 89: 4501–4513.

Yamaguchi

Kwong

Kim

, et al. Phase II study of SMILE chemotherapy for newly diagnosed stage IV, relapsed, or refractory extranodal natural killer (NK)/T-cell lymphoma, nasal type: the NK-Cell Tumor Study Group study. J Clin Oncol 2011; 29: 4410–4416. doi:10.1200/JCO.2011.35.6287.

Pokrovsky

Vinnikov

L-Asparaginase for newly diagnosed extra-nodal NK/T-cell lymphoma: systematic review and meta-analysis.

Expert Rev Anticancer Ther 2017; 17: 759–768. doi:10.1080/14737140.2017.1344100.

Lee

Suh

Park

et al . Extranodal natural killer T-cell lymphoma, nasal-type: a prognostic model from a retrospective multicenter study. J Clin Oncol 2006; 24: 612–618. doi:10.1200/JCO.2005.04.1384.

10.

Querellou

Valette

Bodet-Milin

et al . FDG-PET/CT predicts outcome in patients with aggressive non-Hodgkin’s lymphoma and Hodgkin’s disease. Ann Hematol 2006; 85: 759–767. doi:10.1007/s00277-006-0151-z.

11.

Kostakoglu

Coleman

Leonard

et al . PET predicts prognosis after 1 cycle of chemotherapy in aggressive lymphoma and Hodgkin's disease. J Nucl Med 2002; 43: 1018–1027.

12.

Hutchings

Mikhaeel

Fields

et al . Prognostic value of interim FDG-PET after two or three cycles of chemotherapy in Hodgkin lymphoma. Ann Oncol 2005; 16: 1160–1168. doi:10.1093/annonc/mdi200.

13.

Kim

Choi

Hyun

et al . Risk stratification on the basis of Deauville score on PET-CT and the presence of Epstein-Barr virus DNA after completion of primary treatment for extranodal natural killer/T-cell lymphoma, nasal type: a multicentre, retrospective analysis. Lancet Haematol 2015; 2: e66–e74. doi:10.1016/S2352-3026(15)00002-2.

14.

Harisankar

CN.

Fluoro-deoxy glucose positron emission tomography-computed tomography in a case of natural killer/T-cell lymphoma with bilateral adrenal involvement.

Indian J Nucl Med 2015; 30: 254–255. doi:10.4103/0972-3919.158537.

15.

Fujiwara

Maeda

Nawa

, et al. The utility of positron emission tomography/computed tomography in the staging of extranodal natural killer/T-cell lymphoma. Eur J Haematol 2011; 87: 123–129. doi:10.1111/j.1600-0609.2011.01645.x.

16.

Liu

Zhang

et al . Diagnostic value of 18F-FDG PET/CT for cutaneous extranodal natural killer/T-cell lymphoma, nasal type. Nucl Med Commun 2016; 37: 446–452. doi:10.1097/MNM.0000000000000463.

17.

Wang

et al . Utility of 18F-FDG PET/CT for staging NK/T-cell lymphomas. Nucl Med Commun 2010; 31: 195–200. doi:10.1097/MNM.0b013e32833310fa.

18.

Moon

Cho

Kim

et al . The role of 18F-FDG PET/CT for initial staging of nasal type natural killer/T-cell lymphoma: a comparison with conventional staging methods. J Nucl Med 2013; 54: 1039–1044. doi:10.2967/jnumed.112.113399.

19.

Zhou

Chen

et al . Evaluation of bone marrow involvement in extranodal NK/T cell lymphoma by FDG-PET/CT. Ann Hematol 2015; 94: 963–967. doi: 10.1007/s00277-014-2289-4. Epub 2014 Dec 30.

20.

Cheson

Pfistner

Juweid

et al . Revised response criteria for malignant lymphoma. J Clin Oncol 2007; 25: 579–586.

21.

Kako

Izutsu

Ota

et al . FDG-PET in T-cell and NK-cell neoplasms. Ann Oncol 2007; 18: 1685–1690. Epub 2007 Aug 22. PubMed PMID: 17716987.

22.

Elstrom

Leonard

Coleman

et al . Combined PET and low-dose, noncontrast CT scanning obviates the need for additional diagnostic contrast-enhanced CT scans in patients undergoing staging or restaging for lymphoma. Ann Oncol 2008; 19: 1770–1773. doi:10.1093/annonc/mdn282.

23.

MacDonald

Mulroy

Wilke

et al . PET/CT aids the staging of and radiotherapy planning for early-stage extranodal natural killer/T-cell lymphoma, nasal type: a case series. Radiat Oncol 2011; 6: 182. doi:10.1186/1748-717X-6-182.

24.

Bai

Huang

Cai

et al . Predictive value of pretreatment positron emission tomography/computed tomography in patients with newly diagnosed extranodal natural killer/T-cell lymphoma. Med Oncol 2013; 30: 339. doi:10.1007/s12032-012-0339-0.

25.

Biggi

Gallamini

Chauvie

et al . International validation study for interim PET in ABVD-treated, advanced-stage Hodgkin lymphoma: interpretation criteria and concordance rate among reviewers. J Nucl Med 2013; 54: 683–690. doi:10.2967/jnumed.112.110890.

26.

Khong

Huang

Lee

et al . Midtreatment ¹⁸F-FDG PET/CT scan for early response assessment of SMILE therapy in natural killer/T-cell lymphoma: a prospective study from a single center. J Nucl Med 2014; 55: 911–916. doi:10.2967/jnumed.113.131946.

27.

Lim

Hyun

Kim

et al . Prognostic relevance of pretransplant Deauville score on PET-CT and presence of EBV DNA in patients who underwent autologous stem cell transplantation for ENKTL. Bone Marrow Transplant 2016; 51: 807–812. doi:10.1038/bmt.2016.6.

28.

Suh

Kang

Roh

et al . Prognostic value of tumor 18F-FDG uptake in patients with untreated extranodal natural killer/T-cell lymphomas of the head and neck. J Nucl Med 2008; 49: 1783–1789. doi:10.2967/jnumed.108.053355.

29.

Jiang

Zhang

Jiang

et al . Assessment of the prognostic capacity of pretreatment, interim, and post-therapy (18)F-FDG PET/CT in extranodal natural killer/T-cell lymphoma, nasal type. Ann Nucl Med 2015; 29: 442–451. doi:10.1007/s12149-015-0964-8.

30.

Kim

Hong

Kim

et al . Prognostic value of whole-body metabolic tumour volume and total lesion glycolysis measured on ¹⁸F-FDG PET/CT in patients with extranodal NK/T-cell lymphoma. Eur J Nucl Med Mol Imaging 2013; 40: 1321–1329. doi:10.1007/s00259-013-2443-6.

31.

Song

Chung

Shin

et al . Clinical value of metabolic tumor volume by PET/CT in extranodal natural killer/T cell lymphoma. Leuk Res 2013; 37: 58–63. doi:10.1016/j.leukres.2012.09.011.

32.

Jiang

Kosik

et al . The Deauville 5-point scale improves the prognostic value of interim FDG PET/CT in extranodal natural killer/T-cell lymphoma. Clin Nucl Med 2015; 40: 767–773. doi:10.1097/RLU.0000000000000892.

33.

Liu

et al . Intratumoral heterogeneity of pretreatment 18F-FDG PET images predict disease progression in patients with nasal type extranodal natural killer/T-cell lymphoma. Clin Nucl Med 2016; 41: 922–926. doi:10.1097/RLU.0000000000001375.

34.

Turker

Işık

et al . Nasal natural killer/T-cell lymphoma with skin, eye, and peroneal nerve involvement. Turk J Haematol 2012; 29: 413–419. doi:10.5505/tjh.2012.03360.

35.

et al . Increased body mass index is associated with improved overall survival in extranodal natural killer/T-cell lymphoma, nasal type. Oncotarget 2017; 8: 4245–4256. doi:10.18632/oncotarget.13988.

36.

Kim

Yoon

Jaccard

et al . A prognostic index for natural killer cell lymphoma after non-anthracycline-based treatment: a multicentre, retrospective analysis. Lancet Oncol 2016; 17: 389–400. doi:10.1016/S1470-2045(15)00533-1.

37.

Wang

Shen

Jiang

et al . Prognostic value of baseline, interim and end-of-treatment 18F-FDG PET/CT parameters in extranodal natural killer/T-cell lymphoma: a meta-analysis. PLoS One 2018; 13: e0194435. doi: 10.1371/journal.pone.0194435. eCollection 2018. PubMed PMID: 29558489; PubMed Central PMCID: PMC5860776.

38.

Liang

Ding

Gale

et al . Prognostic value of whole-body SUVmax of nodal and extra-nodal lesions detected by 18F-FDG PET/CT in extra-nodal NK/T-cell lymphoma. Oncotarget 2017; 8: 1737–1743. doi:10.18632/oncotarget.13873.

39.

Jiang

Liu

et al . Predictive approaches for post-therapy PET/CT in patients with extranodal natural killer/T-cell lymphoma: a retrospective study. Nucl Med Commun 2017; 38: 937–947. doi: 10.1097/MNM.0000000000000731. PubMed PMID: 28858180.