Su Liu,Ya-Fei Wang*

1Department of Hematology and Blood and Marrow Transplantation,Tianjin Medical University Cancer Institute and Hospital,National Clinical Research Center for Cancer,Tianjin’s Clinical Research Center for Cancer,Key Laboratory of Cancer Prevention and Therapy,Tianjin 300060,China.

Abstract:

18F-deoxyglucose positron emission tomography/computed tomography(18F-FDG PET/CT)is a functional imaging modality.It has certain advantages over other imaging modalities,such as X-ray,CT,and magnetic resonance imaging,in multiple myeloma(MM).Moreover,both the International Myeloma Working Group and the British Society of Hematology recommend PET/CT as one of the imaging modalities for MM.However,many criteria for the application of PET/CT for evaluating MM in clinical practice have not been determined.Recently,there have been many new advances in the application of PET/CT in MM.This study was designed to review the recent advances in PET/CT for assessing MM.

Keywords:18F-FDG PET/CT;multiple myeloma

Introduction

Multiple myeloma(MM)is a malignant plasma cell tumor in which monoclonal plasma cells proliferate and extensively infiltrate the bone marrow(BM),and peripheral blood and other tissues also infiltrate in the late stage of the disease,accompanied by the secretion of several monoclonal immunoglobulins,with clinical symptoms,including hypercalcemia,renal impairment,anemia,and severe bone destruction [1].18F-deoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT),hereafter PET/CT,is widely used as an imaging modality for diagnosing and staging MM and assessing prognosis,treatment response,and minimal residual disease(MRD).

PET/CT in diagnosing MM

PET/CT is a valuable pretreatment examination tool for patients with newly diagnosed MM(NDMM)or relapsed or refractory MM(RRMM).PET/CT has relatively high sensitivity and specificity in evaluating MM bone destruction and can detect the extramedullary site of proliferating clonal plasma cells,so it plays an important role in diagnosing MM[2].Patients with monoclonal gammaglobulinemia of undetermined significance(MGUS)and smoldering MM(SMM)are managed differently from patients with MM.Furthermore,the management of patients with solitary plasmacytoma(SP)and MM is also different,so accurate diagnosis is particularly important for such patients.In patients with SP of the bone(SPB),as evaluated by conventional imaging modalities,additional unanticipated sites of bone involvement can be detected using PET/CT,increasing the extent of disease and necessitating a revised diagnosis.For these cases,SPB may be a local manifestation of MM where other sites of involvement cannot be detected using other less sensitive imaging modalities(i.e.,skeletal X-ray)or anatomically limited imaging modalities(i.e.,whole-body MRI or CT)[3].Serum free light chain(sFLC)values and the presence of at least two hypermetabolic lesions on PET/CT at diagnosis in patients with SP can help predict their early evolution to MM[4].The 2016 International Myeloma Working Group(IMWG)consensus states that in patients with suspected SP,if whole-body magnetic resonance imaging(WBMRI)is unavailable,whole-body PET/CT must be performed to confirm the diagnosis.Furthermore,in determining the diagnosis of MGUS and SMM,if whole-body X-ray(WBXR)is negative and WBMRI is unavailable,whole-body PET/CT is also required to confirm the diagnosis[2].PET/CT plays an important role in identifying extramedullary lesions,especially rare extramedullary lesions in the liver,testis,prostate,uterus,central nervous system,and muscles and pericardial involvement;cases with positive PET/CT imaging are finally pathologically confirmed to be MM extramedullary lesions[5-11].If patients do not have bone lesions and BM biopsies are negative,PET/CT is valuable in detecting extramedullary myeloma(EMM)[12].

More than 80% of patients with MM have bone involvement at diagnosis,and some studies have compared the value of WBXR,WBMRI,and PET/CT in detecting bone disease in patients with MM.The results showed that PET/CT is more sensitive than WBXR in detecting small lytic bone lesions,whereas it has the same sensitivity as WBMRI in detecting bone disease in the spine and pelvis,and MRI may be superior to PET/CT in diagnosing spinal infiltration patterns[13].The sensitivity of PET/CT in detecting BM involvement in early-stage MM was 90%,and there was a significant correlation between maximum standard unit value(SUVmax)and the plasma cell ratio in BM biopsies(P＜0.01)[14].The combination of PET/CT with low-dose CT can replace the gold standard,traditional means of skeletal examination[15].PET/CT examination with an SUV of more than 3.2 at the fracture site can be used alone in patients with MM to distinguish between old and new pathological fractures.The combination of an SUV of more than 3.5 at diffuse or multiple vertebral lesions and MRI findings in PET/CT examination suggests the possibility of impending fracture[16].A prospective study has compared the number of MM bone focal lesions(FLs)detected using WBMRI and PET/CT and evaluated the diagnostic performance of PET/CT in detecting diffuse infiltrates;the results showed that although WBMRI detected more FLs than PET/CT,no significant difference in the number of FLs detected at the base of each patient was found.PET/CT showed a higher performance in detecting FLs and diffuse infiltrates[17].However,some studies have shown that WBMRI is more sensitive than PET/CT for detecting bone disease in MM[18].Among the whole-body imaging modalities for MM,PET/CT is superior to technetium(99m)sestamibi(Tc-99m MIBI)in detecting FL.However,Tc-99m MIBI is superior in visualizing diffuse disease;among the examinations of spinal and pelvic lesions,MRI is comparable to PET/CT and Tc-99m MIBI in detecting focal and diffuse lesions[19].WBMRI is more sensitive than PET/CT for detecting bone lesions in all regions,except for the skull in patients with NDMM and RRMM,especially for those with a low percentage of plasma cells.WBMRI has sensitivity comparable to that of PET/CT in detecting extramedullary lesions[20].A meta-analysis of 14 studies has assessed the diagnostic accuracy of PET/CT for MM bone and extramedullary lesions and showed that the pooled estimates of sensitivity,specificity,positive likelihood ratio(LR+),and negative likelihood ratio(LR-)of PET/CT for detecting MM extramedullary lesions were 96.0%(95% confidence interval[CI],79.6%-99.9%),77.8%(95%CI,40.0%-97.2%),3.28(95%CI,1.29-8.32),and 0.12(95%CI,0.03-0.42),respectively.The pooled estimates of sensitivity,specificity,LR+,and LR-of PET/CT for detecting MM bone lesions were 61.1%(95%CI,43.5%-76.9%),94.1%(95%CI,71.3%-99.9%),5.73(95%CI,1.53-21.40),and 0.43(95%CI,0.28-0.65),respectively,indicating that PET/CT is a valuable imaging tool for assessing patients with MM,especially for the presence of extramedullary involvement[21].

PET/CT in staging MM

The staging of MM is based on the Durie-Salmon(DS)staging criteria,based on the determination of laboratory parameters,including hemoglobin(Hb),serum calcium(Ga),and serum and urinary M-protein,combined with the presence of skeletal lesions found on conventional imaging modalities [22].Conventional imaging modalities include WBXR and CT.PET/CT is a reliable technique for initially staging patients with MM and has been recommended by the IMWG for staging MM.Recently,PET/CT has become the preferred imaging method for assessing MM due to its technical advantages in sensitivity and specificity;new imaging methods,including PET/CT and MRI of the spine and pelvis,have been integrated into the DS Plus staging system along with relevant laboratory tests[23,24].Studies have evaluated the difference between the DS Plus and DS staging systems or the DS Plus and the revised International Staging System(R-ISS)and showed that the agreement was 0.07 between DS Plus and DS and 0.37 between DS Plus and R-ISS.Multivariate analysis has shown that DS Plus stage III(hazard ratio[HR],11.539;P=0.021)and BM Deauville score(DS)≥4(HR 3.487;P=0.031)were independent prognostic factors of overall survival(OS)in patients with MM and had a better potential in staging and stratifying MM based on the combination of visual description criteria(IMPeTUs)and rISS of PET/CT defined by Italian nuclear medicine experts[25].Based on the results,β2 microglobulin (β2-MG),lactate dehydrogenase,the number of FLs,and SUVmax by PET/CT can predict OS in patients with MM.A new MM staging system(NSS)was further constructed based on the risk coefficients of these four variables;the NSS could better distinguish the risk of ISS stages II and III(P＜0.001 vs P=0.086)[26].However,if PET/CT is selected as the only imaging method,some small lytic skeletal lesions and diffuse spinal involvement[27].

PET/CT in prognosis of MM

SP is a rare plasma cell tumor that can be present in the bone and/or soft tissue and progress to MM.In a study involving 62 patients with SP who underwent PET/CT before any treatment,the prognostic effects of metabolic tumor volume(MTV),total lesion glycolysis(TLG),and other parameters were analyzed.The results showed that standardized uptake fat-free mass was more than 5.2 and standardized uptake body surface area was more than 1.7.In patients with SP at other sites with abnormal uptake on bone SP,PET/CT examination had significantly shorter time for conversion image method to MM[28].Moreover,SMM can progress to MM.A prospective study has shown that positive PET/CT significantly increased the risk of SMM progression to MM.It can be used to identify patients with SMM with a high risk of progression to MM.Furthermore,PET/CT may be a new tool for defining high risk SMM[29].EMM is more commonly seen on PET/CT in the setting of a hematogenous or lymphangitic spread pattern and is less commonly observed as a direct extension of a bone plasmacytoma.The presence of EMM on PET/CT,either in NDMM or during follow-up,is associated with poor prognosis.Patients with EMM,especially those with EMM involving rare sites,such as the liver,lung,and muscle,have a shorter OS[30].PET/CT can help detect extramedullary lesions in patients with NDMM and thus help predict prognosis.The prognosis of patients with EMM was significantly correlated with SUVmax and Ki-67 expression in tissue biopsy samples.The combination of SUVmax and Ki-67 expression as a scoring system could effectively predict progression-free survival(PFS)(P=0.014)and OS(P=0.027)in patients with EMM.Therefore,both SUVmax and Ki-67 expression helped accurately evaluate the prognosis of patients with EMM[31].

Furthermore,PET/CT plays an important role in identifying patients with NDMM[32].Regarding PET parameters,the number of FLs is more likely to be closely related to disease aggressiveness and pathophysiology than SUVmax[33].Both PET/CT and quantitative clonal circulating plasma cells(CPCs)have emerged as emerging tools for predicting the prognosis of MM.The PET-CPC staging system,combining PET/CT with quantitative CPC formation,can facilitate the development of novel risk stratification methods for MM[34].PET/CT can further assist in predicting the prognosis of patients with MM with rISS stages II and III at diagnosis[35].The SUVmax of the bone is a simple and reliable biomarker,and the level of bone SUVmax on baseline PET/CT in MM is closely related to the prognosis of patients[36].Moreover,the MTV on PET/CT images can be used to predict PFS and OS in patients with MM.The total MTV for each patient is defined as the sum of the metabolic numbers of all FLs.Patients with MTV below the cutoff value had significantly better PFS and OS at follow-up than those with MTV above the cutoff value(P＜0.0001)[37].Furthermore,PET/CT can predict the prognosis of patients with NDMM with renal insufficiency(RI).Patients with RI with positive PET/CT had significantly lower survival than those with negative PET/CT;the median PFS was 12.7 and 34.0 months,respectively(P＜0.001),and the median OS was 42.2 months and “not reached,”respectively(P=0.001).Multivariate analysis showed that positive PET/CT was significantly correlated with PFS and OS in patients with RI[38].A retrospective study has assessed the role of IMPeTU parameters in predicting PFS and OS of patients with MM;the IMPeTU parameters were first determined for each patient and compared with metabolic volume parameters used to predict PFS and OS;among the IMPeTU and metabolic volume parameters tested,statistically significant differences were found between SUVmax,MTV,TLG,and the number of lytic bone lesions in patients who progressed(n=26)or did not progress(n=21);PFS and OS were significantly better in patients with MTV≤39.4 mL than those with MTV above the cutoff by the Kaplan-Meier analysis and log-rank test(P=0.0004 and P=0.0001,respectively) [39].The combination of PET/CT and laboratory test data can improve the accuracy of predicting recurrence/progression within 12 months[40].In patients with RRMM,there was a significant correlation between del(17p)/TP53 abnormalities and high SUVmax on PET/CT[41].A study has compared the values of PET/CT,Tc-99m MIBI,and MRI in predicting PFS and OS in patients with MM.The results showed that PET/CT and Tc-99m MIBI may help predict PFS and OS in patients with MM[42].

PET/CT negativity before or after allogeneic hematopoietic stem cell transplantation(HSCT)in patients with MM is a favorable prognostic factor for PFS and OS.High FDG uptake(SUVmax of more than 6.5)is significantly associated with shorter OS than lower SUVmax(＜6.5);furthermore,having more than three FLs before or after HSCT and the occurrence of extramedullary lesions are unfavorable prognostic factors for MM[43].PET/CT negativity after MM treatment is associated with relapse-free survival or long-term disease-free survival.In contrast,a persistently significantly increased SUVmax after treatment is associated with a short time to relapse[44].

A prospective study has analyzed the correlation between PET/CT findings and prognosis in 192 patients with NDMM after thalidomide-dexamethasone (TD)induction therapy and double autologous HSCT(ASCT);the results showed that at least three FLs(44%),SUV of more than 4.2(46%),and extramedullary disease(6%)at baseline adversely affected the estimated 4-year PFS rate;moreover,SUV of more than 4.2 and the presence of extramedullary disease were associated with a lower 4-year OS rate;SUV of more than 4.2 after TD induction and persistence were early predictors of shorter PFS.PET/CT was negative in 65% of the patients 3 months after ASCT,and their 4-year PFS and OS rates were better than those of patients with PET/CT positivity.Multivariate analysis showed that the presence of extramedullary disease,SUV of more than 4.2 at baseline,and persistent positive FDG uptake after ASCT were independent factors for poor PFS.PET/CT examination after induction based on new drugs and subsequent ASCT can be used as a reliable predictor of clinical outcomes[45].The percentage of MM cells in the BM is an important indicator of disease status in patients with MM;increased FDG uptake in BM is related to the percentage of BM plasma cell infiltration in patients with MM.PET/CT may be a useful tool for predicting the level of MM cells in the BM[46].In patients with MM after HSCT,PET/CT may help detect and localize residual and recurrent disease,providing information about the degree of different MM manifestations and the overall disease burden and about the metabolic activity of the disease[47].

PET/CT in response evaluation of MM

PET/CT has shown more rapid changes in imaging findings in patients with MM who respond positively to treatment than contrast-enhanced MRI and is more suitable for treatment efficacy evaluation in MM[48].PET/CT is more specific than WBMRI in detecting MRD in patients with MM after treatment[49].PET/CT is extremely sensitive in detecting intramedullary and extramedullary lesions in MM,and FDG uptake in lesions is related to clinical parameters and plasma cell biological characteristics.PET/CT shows its superiority in assessing treatment response and survival;however,false-positive results may occur when predicting relapse or refractory disease[50].In a study involving 282 patients with MM,treatment response was evaluated using PET/CT;all patients underwent PET/CT at baseline,during posttreatment follow-up,and at relapse.The results showed that 42% of the patients had more than three FLs,50% had SUVmax of more than 4.2,and 5% had extramedullary disease at the initial diagnosis.Approximately 70% of the patients achieved PET/CT negativity after treatment and had a favorable PFS and OS.In patients achieving conventionally significant CR,PET/CT negativity was an independent predictor of prolonged PFS and OS,and SUVmax of more than 4.2 after first-line treatment and persistence was independently associated with MM disease progression[51].Case reports have suggested that PET/CT is superior to MRI in assessing early response to monoclonal antibodies,immunomodulatory drugs,and dexamethasone in patients with MM[52].After treatment,both PET/CT and WBMRI provide information on the extent of MM response.PET/CT provides a more comprehensive assessment of progressive or relapsed MM;therefore,PET/CT may be more appropriate than MRI for determining disease response status in patients with MM after treatment[53].

Interpretation criteria for PET/CT in MM

The IMWG guidelines recommend PET/CT for assessing MM treatment response;however,simple and reproducible interpretive criteria are still lacking.Recently,Italian nucleologists have proposed IMPeTUs to standardize the assessment criteria of PET/CT in patients with MM[54].This criterion was based on data from a multicenter clinical trial(EMN02),which included 86 patients with MM who underwent PET/CT at baseline(PET0),after induction therapy(PET-AI),and at the end of treatment(PET-EoT).The patients were prospectively tested for blinded independent central review according to the IMPeTUs criteria.Moreover,all evaluators were tested for agreement on positive or negative PET/CT scans for different positive thresholds(i.e.,DS2,DS3,DS4,and DS5),and the main parameters included BM,limited lesion,and extramedullary disease scores.The results showed that the percentage agreement was better than 75% at all time points,and there was 100% agreement in assessing the presence of skull lesions after treatment,and comparable results were obtained at any time point(i.e.,PET0,PET-AI,or PET-EoT)when concordance analysis was performed using Krippendorff’s alpha coefficient.DS was demonstrated to be highly reproducible,being more pronounced in patients with a score of 4.However,whether the IMPeTUs criteria can be a judgment criterion for the application of PET/CT in MM needs validation by prospective clinical studies based on the IMPeTUs criteria.

PET/CT should take precedence over other imaging techniques in assessing the response status after ASCT.The use of PET/CT may help detect single or multiple non-secretory areas of recurrence early in patients with non-secreting MM.Furthermore,PET/CT and contrast-enhanced MRI may better define the osteonecrosis of the jaw,its location,and its extent in patients with MM treated with bisphosphonates[55].Some researchers have proposed a quantitative method for assessing the bone and BM using PET/CT in MM.Therefore,the degree and intensity of bone FDG uptake,that is,bone participation intensity(IBI),and the percentage of bone involvement(PBI)should be calculated and considered.IBI is defined as the PBI multiplied by the mean SUV higher than liver uptake.Visual analysis divided PET examinations into three groups(i.e.,negative/mild,moderate,and overt bone involvement),which had different IBI ranges(P＜0.0001).The results showed that the IBI score was an objective measure of bone and BM involvement in MM,allowing patients to be classified into different degrees of bone disease aggressiveness[56].

Investigators have analyzed two independent European randomized phase III trials (i.e.,IFM/DFCI2009 and EMN02/HO95)for transplant-eligible patients with NDMM to determine the criteria for posttreatment PET complete metabolic response(CMR).Lesion status assessment before and after treatment was assessed using PET/CT according to the DS5 scoring method.The results showed that at baseline,78% of the patients had FL lesion FDG uptake score(FS)(11% extramedullary)and 80% had FS of≥4;all patients had diffuse uptake in the BM(35.5% had BMS of≥4);after maintenance therapy(PM),31% of the patients had macroscopically detectable FLs(2% extramedullary),of which 24% and 67.7% had FS of 3 and≥4,respectively.During PM,98% of the patients had residual BM diffuse uptake that was significantly lower than that at baseline(mainly BMS of 2-3;only 8.7% had BMS of≥4).Univariate and multivariate analyses showed that FS and BMS of less than 4 at PM are associated with prolonged PFS and OS,suggesting that FL and BM FDG uptake below the hepatic background after MM treatment are independent predictors of improved PFS and OS and can be used as a criterion for achieving CMR using PET in patients with MM[57].

Novel radioisotope PET/CT in MM

FDG is the most widely used tracer but has relatively low sensitivity in detecting MM.The use of novel radioisotopes,such as 11C-methionine(11C-MET)and 68Ga-pentixafor,inPET/CT can provide more information related to MM disease.

The results of a study assessing 10 patients with SMM have shown that PET/CT with 11C-MET and 68Ga-pentixafor showed higher sensitivity than18F-FDG PET/CT in detecting BM involvement in SMM[58].A prospective cohort study[59]has compared the performance of 68Ga-pentixafor and18F-FDG PET/CT in newly diagnosed MM and showed that the positive rate of 68Ga-pentixafor PET/CT was higher than that of18F-FDG PET/CT in patients included in the study(93.3% vs 53.3%;P=0.0005);quantitative analysis has shown that the BM uptake value of 68Ga-pentixafor(i.e.,TBmUCXCR4,SUVmax,and SUVmean)was significantly correlated with end-organ damage,stage,and laboratory biomarkers associated with tumor burden,including serum β2-MG,sFLC,and 24-h urinary light chain levels(P ＜ 0.05).11C-MET and 11C-4'-thiothymidine(11C-4DST)C-MET PET/CT are more useful than18F-FDG in detecting active lesions in MM,especially in the early stages of the disease[60].11C-acetate PET/CT is more accurate than18F-FDG PET/CT in diagnosing MM,identifying patients,and monitoring treatment response[61].18F-fluoro-ethyl-tyrosine(18F-FET)PET/CT is more accurate in identifying bone disease and can replace18F-FDG PET/CT for diagnosing MM-related bone disease[62].The combined use of FDG and fluorothymidine PET/CT in MM represents a valuable prognostic indicator for identifying patients at a higher risk and may help stratify patients suspected of relapse[63].No significant difference in the number of lesions detected by gallium-68-labeled DOTA-TATE (Ga-TATE)PET/CT was found between the two modalities in patients with MM(P=0.67);however,diffuse BM uptake of TATE appeared to be a predictor of OS(P=0.033);TATE PET/CT appeared to be an alternative imaging modality to18F-FDG PET/CT and may play a complementary role in MM management[64].For patients with RRMM,choline-based tracers may overcome some limitations of18F-FDG,including the lack of sensitivity in describing skull lesions and the fact that 10% of patients with MM are negative for FDG uptake;choline-based tracers can additionally detect as many as 75% of lesions[65].Some investigators have compared FDG with the newer metabolic tracer18F-fluorocholine(FCH)and detected MM lesions when the disease recurred or progressed;the results showed that PET/CT examinations performed for suspected recurrent or progressive MM,when using FCH,showed more lesions[66].

Summary

PET/CT plays an important role in diagnosing and staging MM and stratifying the prognosis and evaluating treatment efficacy in MM(Figure 1).The findings of this study will help solve few puzzles for physicians and help them use PET/CT better in MM,thus,helping physicians manage patients with MM better.

Figure 1 The role of PET/CT in MM.

Future

There are still some unmet clinical needs in the clinical application of PET/CT in MM,among which the most important is the lack of unified interpretation criteria,mainly including the criteria for positive different types of lesions(i.e.,bone,BM,extramedullary)before treatment and the timing and criteria for efficacy judgment after treatment.Although some studies have explored the feasibility of the DS method,they have not formed a systematic conclusion,and the DS method is expected to be unified interpretation criteria in MM application but needs more studies to confirm.In the future,the use of novel radioisotopes is expected to improve the clinical application value of PET/CT in MM.