Prognostic and predictive value of nuclear imaging in endocrine oncology

Abstract

In the last few years, the role and use of medical technologies in (neuro)endocrine oncology has greatly evolved allowing not only important diagnostic information but also prognostic stratification in different clinical situations. The terms “prognostic” and “predictive” are commonly used to describe the relationships between biomarkers and patients’ clinical outcomes but have quite different meaning. The present work discusses the prognostic and predictive value of nuclear medicine imaging. It critically reviews the clinical significance and potential impact of molecular examinations on follow-up and therapeutic strategies in patients with neuroendocrine neoplasms, thyroid tumors, and adrenal malignancies.

Introduction

Patient’s prognosis is usually defined as “the prediction of the likely outcome of an illness based on the individual’s condition and the usual course of the illness as seen in similar situations” [1–3]. Prognosis should not be confused with prediction [4]. However, the terms “prognostic” and “predictive” are commonly used interchangeably to describe the relationships between biomarkers (i.e.: imaging) and patients’ clinical outcomes. Determining prognostic factors require prospective or retrospective studies correlating a clinical parameter with the characteristics of the patient and the stage of the disease. According to Clark et al. [2], a prognostic factor is a measure associated with clinical outcomes in the absence of treatment or after standard treatment and reflects the natural history of the disease. On the other hand, a predictive factor is an index that reflects the chance of response to a particular treatment that can be defined using any of the clinical parameters used in clinical studies. For example, HER2-overexpression is a prognostic factor of poor survival on retrospective tissue analysis from series of breast cancer patients treated with chemotherapy. However, HER2-overexpression is also predictive of response to HER2-directed therapies. Moreover, with the generalized use of HER2-targeted therapies the prognosis has changed and currently HER2-overexpression has rather become a prognostic factor for better outcome than for some other breast cancer subtypes. Overall, a predictive factor is an interaction between the benefits of treatment and the state of the biomarker that is better assessed in randomized trials with a control group. Beyond the expected clinical benefits of personalized medicine, theragnostics could also have a significant positive economic effect. For example, the implementation of KRAS mutation status as a companion test to exclude patients unlikely to benefit from such therapies has been estimated to a net cost of $7500–12,400 per patient in the United States and €3900–9600 per patient in Germany, with equivalent clinical outcomes [5].

One of the main difficulties in establishing the prognostic role of new imaging modalities is what could be called as “staging bias”. For instance, if a more sensitive imaging modality becomes available, it may have an impact on tumor staging by modifying it radically (i.e., localized vs. metastatic disease). In these cases, it is essential to determine if staging modification that is only related to the use of new imaging techniques justifies a modification of therapeutic strategy. This is a possible situation in nuclear medicine imaging where the development of new radiotracers and technological solutions (i.e. scanning protocols, technological advances) may result in the detection of an increased number and sites of lesions without final proof that these findings will have an impact on treatment strategy and patient outcome. However, there is a growing spectrum of data regarding the usefulness of functional imaging techniques to define patient prognosis. Accordingly, the present work emphasizes the prognostic and predictive value of nuclear medicine investigations in endocrine oncology with a focus on gastroenteropancreatic (GEP) neuroendocrine neoplasms (NENs), thyroid tumors, and adrenal malignancies.

GEP-NENs

Somatostatin receptor-based imaging

GEP-NENs are heterogeneous tumors in terms of their molecular biology, clinical behavior, prognosis, and response to therapy [6]. The majority of them express somatostatin receptors (SR) on cell surface, which is the rationale for the use of somatostatin analog (SSA)-based diagnostic imaging and therapies in this clinical situation [7]. SR expression is a prognostic factor itself in NENs being associated with longer OS [8]. SR expression levels vary considerably depending on the location of the primary tumor [9]. For instance, lower SR expression levels were found in NENs originating from the appendix or colon, compared with NENs from other anatomic origins. Interestingly, NENs derived from appendix or colon are associated with significantly worse patient outcomes [9]. These differences of SR expression, which are strictly related to patient prognosis, should be taken into account when SR-based imaging is used as diagnostic or prognostic marker. Moreover, some histological subtypes of NENs are characterized by a low expression of SR requiring the use of different radiotracers beyond radiolabeled-SSA. As a matter of fact, for insulinomas ¹⁸F-fluorodihydroxyphenylalanine (¹⁸F-FDOPA) or tracers targeting the glucagon-like peptide-1 receptors have to be preferred [10]. Similarly, among well-differentiated lung NENs, atypical carcinoids are characterized by lower SR expression then typical carcinoids, therefore they are better detected by using ¹⁸F-FDG rather than SR-based imaging. Therefore, the use of dual-tracer PET/CT (¹⁸F-FDG and radiolabeled-SSA) could provide useful diagnostic and prognostic information in well-differentiated lung NENs, in particular when the histological subtype (typical vs. atypical carcinoid) is unknown [11].

The integration of diagnostics and therapeutics (ther-agnostics) represents a major opportunity to detect the disease early and to select appropriate treatment, monitor treatment, and determine prognosis. Therapy with a long acting cold SSA is highly effective in controlling symptoms related to tumor secretion. Moreover, SSAs are also considered to have antiproliferative effects [12, 13]. Peptide receptor radionuclide therapy (PRRT) targeting the SR is an excellent illustration of patient-specific therapy based on the image-and-treat approach. In the context of NENs, it means the use of molecular vectors labeled either with diagnostic or with therapeutic radionuclides to form a theragnostic couple (i.e. 68Ga-DOTATATE and 177Lu-DOTATATE). Interestingly, in patients with NENs, treatment with “cold SSA” may not always require evidence of tumor’s radio-tracer uptake on somatostatin receptor scintigraphy (SRS) or positron emission tomography-computed tomography (SR-PET/CT) to be prescribed. On the other hand, significant tumor uptake on SRS/SR-PET/CT is mandatory before PRRT with “hot SSA”. Indeed, both SRS and SRPET/CT with 68Ga-labeled-SSA are functional imaging providing data about the density of SR subtype-2 in NENs [14]. These imaging methods have been proposed as potential predictors of response to SSA-based therapies [7]. The correlation between SSA-based imaging and response to SSA-based treatment has been demonstrated by several studies showing a better response to SSA-based therapies in NENs with high radiolabeled SSA uptake in tumor lesions [15–18]. In addition, SRS may underestimate disease extension in organs with higher radiotracer physiological uptake (e.g., liver and gut) and in small volume disease, given gamma camera’s low spatial resolution [19]. SR-PET/CT has several advantages compared with SRS: the higher spatial resolution of SR-PET/CT compared with SRS allows the detection of more and/or smaller NEN lesions, or with low to moderate SSTR expression, resulting in a higher sensitivity and diagnostic accuracy. Therefore, SR-PET/CT is usually proposed to assess which patients might benefit from PRRT. Moreover, SR-PET/CT offers straightforward quantitation, which potentially allows for a more robust patient selection than visual assessment based on SRS [7, 17]. Overall, a low radiolabeled SSA uptake was reported to be associated with poor prognosis in patients with NENs [20]. One possible explanation of this finding is the lower expression of SSTRs usually due to poor differentiation as observed in high-grade NENs [20]. The uptake intensity of ⁶⁸Ga-labeled-SSA evaluated on baseline PET/CT can be useful to estimate the delivered absorbed dose on tumor during PRRT and to predict the response to SSA-based therapies. Ezzidin et al. evaluated 61 lesions in 21 patients treated with ¹⁷⁷Lu-DOTATATE showing positive correlation between pre treatment ⁶⁸Ga-DOTATOC SUV-max or SUVmean and the tumor absorbed dose during the first cycle of PRRT [21]. Lesions with high SUV values (SUVmax > 25; SUVmean > 15) received a dose of more than 10 Gy/GBq in two thirds of lesions and a dose of less than 5 Gy/GBq in only around 10% of cases. All lesions with low SUV values (SUVmax < 9; SUVmean < 7) had a dose less than 5 Gy/GBq. PRRT benefit could be determined not only by whether the tumors take-up the radio-pharmaceutical, but also high uptake can be considered predictive of greater benefit [7]. Pre-therapeutic uptake on ⁶⁸Ga-DOTATOC was highly predictive of response at 3 months after ⁹⁰Y-DOTATOC PRRT [22]. All responders (n = 20) had an SUVmax higher than 17.9 whereas 15 out of 16 non-responders (94%) had a SUVmax lower than this threshold. Others have found a SUVmax above 16.4 on ⁶⁸Ga-DOTATOC PET/CT before PRRT (90Y-DOTATOC and/or ¹⁷⁷Lu-DOTATATE) as a sensitive predictor (sensitivity: 95%) for lesion stabilization or shrinkage in 60 liver metastases in 30 patients [17].

Several PET parameters have been suggested to quantify tracer uptake by the tumor, as measurement of tumor uptake intensity (i.e. SUV), tumor-to-liver, and tumor-to-spleen SUVmax ratios [17, 20]. A structured reporting system for SR-PET/CT, has also been proposed as a standardized assessment for diagnosis and treatment planning in patients with neuroendocrine tumors [23]. In addition to the conventional PET parameters, computational tools that enable automatic or semiautomatic measurement of disease burden based on radiolabeled SSA-avid tumor volume might help us to better predict disease outcomes [20]. Furthermore, textural tumor features approximate intratumoral SR heterogeneity. In exploratory analyses, they have been found to be superior to the aforementioned conventional parameters for predicting the survival of patients with NENs [20].

However, no final consensus has been reached concerning the predictive value of SR-PET/CT to assess treatment response [16]. The issue is further complicated because (i) there are no well-validated cut-off criteria to assess response to SSA-based treatment using baseline SRPET/CT and (ii) because the several available SSAs have different affinities among the 5 SRs subtypes and different physiological distributions. Moreover, any cut-off identified by retrospective data needs further validation. Additional prospective studies are warranted to define thresholds for semi-quantitative PET parameters below which the probability of benefit from SSA-based therapy is sufficiently low to refrain from treatment [7].

¹⁸F-fluorodeoxyglucose PET/CT imaging

In oncology, 18F-fluorodeoxyglucose (¹⁸F-FDG), a radiolabeled glucose analog has emerged as a chief radio-pharmaceutical for the PET/CT investigations. The role of ¹⁸F-FDG PET/CT in low-grade NENs is limited due to low metabolic activity and slow growth of these tumors. However, ¹⁸F-FDG PET/CT becomes a preferred modality in the staging of NENs with high Ki67 labeling index and/or poorly differentiated tumors (G2 tumors with high Ki67 labeling index, G3 tumors, and neuroendocrine carcinoma (NEC)) [24, 25]. In patients with NENs, prognostic strati-fication can potentially be achieved by ¹⁸F-FDG PET/CT. In recently published studies, intensity of ¹⁸F-FDG uptake in NENs has been reported to be inversely related to the patient survival [26, 27]. Therefore, an aggressive behavior with less favorable long-term survival was found in NENs with increased ¹⁸F-FDG uptake (Fig. 1) [26, 27]. Moreover, ¹⁸F-FDG uptake is correlated to the proliferative index, which is also associated with tumor response to chemotherapy [28]. However, the correlation between ¹⁸F-FDG uptake and response to chemotherapy has not yet clearly been established.

Fig. 1.

A 54-year-old patient with bone and liver metastasis (dotted arrows) of pancreatic neuroendocrine carcinoma (Ki67: 80%; arrows) presenting with anorexia and weight loss. Patient underwent ⁶⁸Ga-DOTATOC (a, c) and ¹⁸F-FDG (b, d) dual-tracer PET/CT investigation for patient prognostication and selection before 177Lu-DOTATATE peptide receptor radionuclide therapy (PRRT). Primary pancreatic tumor (arrows) and metastatic disease (dotted arrows) showed preponderant and sometimes exclusive 18F-FDG uptake accordingly with pathological features of aggressiveness

In early prediction of rapidly progressive well differentiated NENs, ¹⁸F-FDG PET/CT was found to be more accurate than Ki67. Among SR positive patients, a positive ¹⁸F-FDG PET/CT enables to accurately stratify patients with poorer survival [26, 27]. According to these findings, a metabolic grading based on 18F-FDG PET/CT was proposed in patients with metastatic GEP-NENs suggesting ¹⁸F-FDG PET/CT as an effective and noninvasive technique to predict patient’s overall survival (OS) in complement with the conventional Ki67 [29].

¹⁸F-FDG PET/CT provides whole body investigation, allowing in vivo extensive tumor characterization that is potentially useful for guiding biopsy with consequent influence on therapeutic strategy aggressiveness [30]. However, the presumed value of PET/CT for guiding biopsy-based proliferation assessments in metastatic disease should be further investigated.

Finally, 18F-FDG PET/CT could also predict PRRT response, allowing the selection of grade 1 and 2 metastatic NENs patients who might benefit from more intensive treatment protocols [31, 32].

Dual-tracer PET/CT imaging

Ideally, to increase the detection of NENs a dual-tracer PET/CT strategy (performing both SR-imaging and ¹⁸F-FDG PET/CT) could be suggested because these radiopharmaceuticals provide complementary information [24, 33, 34]. Flip-flop effect is the phenomenon usually associated with high-grade NENs where a pattern of high ¹⁸F-FDG uptake is observed along with low radiolabeled SSAs and/or 18F-FDOPA uptake (Fig. 1). Conversely, a pattern of low 18F-FDG uptake with high radiolabeled SSAs and/or 18F-FDOPA uptake is observed in low-grade NENs [19, 24, 35, 36] (Fig. 2). However, a wide overlap occurs in this regard and even low grade NENs can be 18F-FDG avid. However, dual-tracer PET/CT is not always feasible because of financial constraints, and the choice of PET radio-pharmaceutical should reflect both patient clinical setting and tumor pathological features of aggressiveness [24]. Accordingly, a value of Ki67 above or equal to 10% is usually but empirically considered as the cut-off to propose ¹⁸F-FDG PET/CT for well-differentiated NENs [37]. Of course, SR-imaging remains mandatory before PRRT.

Fig. 2.

Typical example of “flip-flop” effect on metabolic imaging in a 57-year-old patient with voluminous and locally infiltrating pancreatic G2 NEN (Ki67: 10%). Patient underwent ⁶⁸Ga-DOTATOC and ¹⁸FFDG dual-tracer PET/CT investigation for prognostication and selection before ¹⁷⁷Lu-DOTATATE peptide receptor radionuclide therapy (PRRT). Pancreatic mass showed intense ⁶⁸Ga-DOTATOC uptake (a, c; arrow) and almost absent ¹⁸F-FDG uptake (b, d; dotted arrow)

Dual-tracer PET/CT approach has also been suggested for assessing the potential benefit that a patient with NEN may receive from PRRT [20, 33, 38]. Chan et al. [34] performed a retrospective study of patients with metastatic NENs that underwent both SR-PET/CT and ¹⁸F-FDG PET/CT, developing a score based on the lesion uptake of two different radiopharmaceuticals (the ‘NETPET score’). The score was associated with patients’ survival in a multivariate analysis. However, except for PRRT, which require uptake on SR-imaging, none of the current nuclear imaging has proved its value in predicting response to chemotherapy or targeted therapies. A patient with significantly positive ¹⁸F-FDG PET/CT will comparatively have a shorter PFS after PRRT than a patient with negative ¹⁸F-FDG imaging, but this does not necessarily mean that the patient would not derive benefit from PRRT. Thus, prospective randomized trials in this regard are needed.

¹⁸F-fluorothymidine (¹⁸F-FLT) PET/CT has also been investigated as a prognostic marker for NENs in comparison and associated with ¹⁸F-FDG PET/CT and Ki67 index. A multivariate model revealed that ¹⁸F-FLT PET only adds information regarding progression free survival (PFS) but not OS, whereas ¹⁸F-FDG PET remains predictive of both PFS and OS. Accordingly, the role of 18F-FLT PET in NENs remains to be established [39].

Although 18F-FDOPA PET/CT is accurate in patients with small intestinal NENs [14], at present it has not been used to predict the efficacy of antitumoral treatment.

Thyroid tumors

Radioiodine imaging in differentiated thyroid carcinoma

Iodine-131-based radioiodine therapy is the mainstay for both the postoperative adjuvant treatment of intermediate to high risk differentiated thyroid carcinoma (DTC) and therapy of locally advanced and metastatic DTC [40, 41]. Two-thirds of patients with metastatic DTC show radioiodine (RAI) uptake and complete remission is obtained in one-third of patients after RAI therapy while long term stabilization and symptoms control is often generally achieved in remaining ones [42, 43].

Patients having DTC with no RAI uptake [i.e. RAI-Refractory (RAIR) DTC] have a worse prognosis compared with patients with RAI-avid thyroid tumors [19]. It has been recently reported that diagnostic Iodine-124 PET/CT may be highly accurate in predicting findings on post-therapy RAI scanning and this method was shown to have a high prognostic power in patients with DTC [42].

¹⁸F-FDG and ¹²⁴I PET/CT imaging in differentiated thyroid carcinoma

The clinical significance of information provided by ¹⁸F FDG PET/CT and its impact on DTC management is still debatable amongst experts. So far, few studies in patients with DTC have evaluated the prognostic role of ¹⁸F-FDG PET or PET/CT performed at different times, particularly in predicting disease persistence or progression and/or survival [43, 44]. An intense ¹⁸F-FDG uptake of the primary DTC, initially detected by PET/CT as thyroid incidentaloma is found to be associated with persistence/progression of disease. However, on a multivariate analysis of all other prognostic factors, ¹⁸F-FDG uptake alone did not add any further prognostic information. Nonetheless, the information regarding conventional predictors could only be obtained following surgery [45].

Overall, based on the available literature, there exists significant data suggesting the prognostic role of ¹⁸F-FDG PET/CT in DTC patients both after primary surgery and at restaging, as 18F-FDG PET result was found to be an independent prognostic factor by several studies [46]. Thus,

¹⁸F-FDG based imaging is helpful in identifying DTC patients with higher risk of developing distant metastases or patients with distant metastases at higher risk of developing disease progression [19, 46, 47]. An intense 18F-FDG uptake in thyroid tumors is often associated with resistance to radioiodine and consequently to a worse patient survival [48]. ¹⁸F-FDG PET/CT could be proposed in selected high-risk patients (e.g., patients with rising thyroglobulin and negative iodine-131 imaging) as a powerful method to detect surgically resectable disease for the selection of patients who may benefit from curative surgery. Furthermore, because of its prognostic value, the aggressiveness of therapy for high-risk DTC patients with metastases could be modulated based on ¹⁸F-FDG PET/CT positivity [43, 49].

¹⁸F-FDG PET/CT is a powerful investigation to recognize patients at high risk of disease progression after RAI therapy. ¹⁸F-FDG PET/CT is able to identify nodal metastases or systemic spread not or partially detected by Iodine-131 whole body scan (WBS). Moreover, in advanced DTC, ¹⁸F-FDG PET/CT may detect new RAI-negative metastases in patients with increasing Tg levels but with unchanged positive Iodine-131 WBS [50]. In both clinical scenarios, no RAI approach could be proposed but alternative treatments should be considered.

¹⁸F-FDG PET/CT findings could guide the decision-making process particularly in patients with aggressive histological subtypes of thyroid cancer, showing aggressive clinical behavior with high glucose metabolism and intense ¹⁸F-FDG uptake [19, 51]. BRAFV600E mutation and high ¹⁸F-FDG uptake are considered as potential prognostic factors in patients with papillary thyroid cancer (PTC). Interestingly, the presence of the BRAFV600E mutation has been independently associated with high 18F-FDG uptake on preoperative PET/CT in patients with PTC [52]. In summary, based on current evidences, the role of ¹⁸FFDG PET/CT is primarily supported in patients with high-risk DTC with elevated serum thyroglobulin under stimulation (generally > 10 ng/mL) after primary treatment and negative RAI imaging [53]. ¹⁸FDG PET/CT scanning may also be considered as (i) a part of initial staging in poorly differentiated thyroid cancers and invasive Hürthle cell carcinomas, (ii) a prognostic tool in patients with metastatic disease to identify lesions and patients at highest risk for rapid disease progression and disease-specific mortality, and (iii) an evaluation of post-treatment response following systemic or local therapy of metastatic or locally invasive disease [54].

¹²⁴I PET/CT could be used in clinical practice for diagnostic purposes and dosimetry assessment before RAI therapy in patients with DTC [55]. ¹²⁴I PET-based dosimetry allows the administration of high therapeutic activities and the identification of patients at higher risk of developing hematological toxicity after RAI therapy [56]. At the same time, the identification of iodine-avid DTC localizations on ¹²⁴I PET/CT could improve the selection of patients who will benefit from RAI treatment, avoiding unnecessary RAI administrations. However, controversial results have been reported in this setting [57, 58].

PET/CT imaging in medullary thyroid carcinoma (MTC)

Currently, ¹⁸F-FDOPA seems to be the most useful PET radiopharmaceutical in detecting recurrent MTC based on rising levels of serum calcitonin [59]. Moreover, a predictive and prognostic value of ¹⁸F-FDOPA PET/CT over conventional imaging has been recently suggested in patients affected by recurrent MTC. Accordingly, PFS and disease-specific survival (DSS) result significantly longer in patients with an unremarkable ¹⁸F-FDOPA PET/CT compared with those with positive studies [60].

In patients with a more aggressive disease, ¹⁸F-FDG is usually preferred for PET imaging in detecting MTC recurrence (Fig. 3) [61]. In fact, ¹⁸F-FDG PET/CT is able to distinguish progressive from stable MTC disease. Moreover, the combined use of ¹⁸F-FDG PET/CT and calcitonin doubling time seems to have a prognostic value improving the identification of high-risk MTC patients for which a close monitoring is required [62].

Fig. 3.

A 64-year-old woman with metastatic medullary thyroid carcinoma presenting with highly increased serum ACE level. Patient underwent ¹⁸F-FDG (a, c, e) and ¹⁸F-FDOPA (b, d, f) PET/CT for primary staging and prognostication, showing preponderant tumor ¹⁸F-FDG uptake. Note of worth, hepatic metastases significantly progressed between two PET/CT examinations performed at 1-month interval, confirming aggressive CMT phenotype

Several articles demonstrated that ¹⁸F-FDG PET is an independent predictive factor related to the response to pre-targeted radioimmunotherapy [63] or tyrosine kinase inhibitors (TKI) as vandetanib [64, 65]. The metabolically most active lesion at ¹⁸F-FDG PET/CT can predict progression free survival (PFS) in patients with MTC starting treatment with TKI; however, this metric failed in OS prediction [64, 65]. Recently, intra-tumoral textural features as well as volumetric parameters derived by pre-therapeutic ¹⁸F-FDG PET, such as the total lesion glycolysis (TLG), have been recently demonstrated as independent parameters for OS prediction in metastatic MTC patients [64], suggesting a possible role of radiomic approach to MTC.

About somatostatin receptor PET, currently, there is not clear evidence about its possible prognostic value in patients with MTC [66].

Adrenal malignancies

PET/CT imaging in adrenocortical carcinoma

Adrenocortical carcinoma (ACC) is a rare and deadly malignancy with few recognized prognostic factors. ¹⁸FFDG PET is of interest for the management of ACC in several clinical situations such as diagnosis, primary staging, and relapse detection during postsurgical follow-up. ¹⁸F-FDG PET/CT is useful in differentiating benign from malignant adrenal masses allowing the exclusion of adrenocortical carcinoma diagnosis with high negative predictive value [67]. ¹⁸F-FDG PET has been challenged for prognostic stratification in patients with ACC showing contradictory results [68, 69]. However, according to recent evidence, ACC patients with high whole-body metabolic tumor volume, TLG, and SUVmax have a worse prognosis and OS, suggesting that measurement of metabolic parameters on 18F-FDG may be helpful for optimizing the therapeutic strategy [70].

PET/CT imaging in pheochromocytoma and paraganglioma

For pheochromocytomas (PHEOs) and paragangliomas (PGLs), imaging phenotype is tightly linked to the tumor genotype [71]. Around 5–10% of solitary PHEOs are hereditary, whereas tumor multiplicity or extra-adrenal tumors are related to currently known germline/somatic mutations in 40–70% of patients. SDHx genes comprise the SDHA, SDHB, SDHC, and SDHD genes, which encode the four subunits of succinate dehydrogenase (collectively called SDH, also known as the mitochondrial complex II. PHEO/PGLs with an underlying SDHB mutation are associated with a higher risk of aggressive behavior leading to development of the metastatic disease. Importantly, mutation of any of the SDH genes leads to the disruption of the TCA cycle with subsequent accumulation of succinate [72]. In SDHx-related tumors, succinate is thought to be one of the important determinants of molecular imaging phenotype that typically consists of high 18F-FDG uptake and low ¹⁸F-FDOPA or ¹²³I-metaiodobenzylguanidine (¹²³I-MIBG) uptake (Fig. 4). Taken together, this suggests that imaging phenotype should be used as a genotypic marker rather than a prognosticator [73]. However, no study has so far evaluated the prognostic value ¹⁸F-FDG or ¹⁸F-FDOPA-derived metabolic indices in metastatic PHEO/PGL of various genotypes [74].

Fig. 4.

Typical molecular imaging phenotype of SDHB-related abdominal paraganglioma (PGL) (a) in a 44-year-old patient. Both PGL (arrow) and para-aortic nodal metastasis (dotted arrow) showed no ¹²³I-MIBG uptake (b) but high ¹⁸F-FDG uptake (c, d). In patients with PGL this imaging pattern should be used as a genotypic marker rather than a prognosticator

Future perspectives and concluding remarks

The use of nuclear imaging in endocrine oncology for disease prognostication is constantly evolving. The integration of diagnostics and therapeutics (i.e. theragnostic) by in vivo molecular imaging represents a major opportunity to select appropriate treatment and to determine prognosis in endocrine malignancies. In some cases, this might require more than one molecular imaging probe with complementary information for better characterization of the disease.

In patients with well-differentiated NENs SR-PET/CT is required before PRRT, but no final consensus has been reached concerning the predictive value of SR-PET/CT in assessing treatment response. 18F-FDG PET/CT offers additional prognostic information in these patients. Hence, a dual-tracer PET/CT imaging with 68Ga-SSA and ¹⁸F-FDG provides an attractive opportunity for non-invasive characterization of disease phenotype and for refining the therapeutic strategy. Indeed, several therapeutic options are nowadays available. However, the optimal therapeutic strategy and sequencing of available therapies in a given patient has not been yet completely defined. Toxicity and quality of life are also important elements in the therapeutic decision and must be considered for each single patient. Thus, the early identification of non-responders to a specific option will improve the long-term patient management by providing a personalized treatment and lowering the iatrogenic toxicity. In metastatic DTC, the role of ¹⁸F-FDG PET/CT is well accepted as it offers complementary information to radioiodine imaging that can be helpful to refine prognosis and predict outcome. In relapsing MTC, ¹⁸F-FDOPA PET/CT seems to have a predictive and prognostic value, as well as ¹⁸F-FDG PET/CT (in association with calcitonin doubling time) in patients with more aggressive disease. Some recent exploratory investigations in NENs and MTC, revealed the interest of tumor textural features assessment. This kind of analysis, eventually performed from multi tracer PET investigations, could offer in case of adrenal malignancies a non-invasive and in vivo representation of metabolic tissular heterogeneity and different tumor biological behavior. Radiomics applied to PET metabolic cartography provides individual quantitative and objective information that could be combined with clinical and genomic features, in diseases with important phenotypic and genomic substrates as PHEO and PGL. The PET radiomic approach obtained before and/or during treatment could represent a new and exciting research direction and may be used for predicting patient survival or to identify tumors with high metastatic potential, particularly in cases without metastatic spread at diagnosis.