New imaging approaches to phaeochromocytomas and paragangliomas

Authors

  • Bas Havekes,

    1. Department of Internal Medicine, Division of Endocrinology, University Hospital Maastricht, Maastricht, The Netherlands
    2. Reproductive and Adult Endocrinology Program, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD, USA
    Search for more papers by this author
  • Kathryn King,

    1. Reproductive and Adult Endocrinology Program, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD, USA
    Search for more papers by this author
  • Edwin W. Lai,

    1. Reproductive and Adult Endocrinology Program, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD, USA
    Search for more papers by this author
  • Johannes A. Romijn,

    1. Department of Endocrinology and Metabolism, Leiden University Medical Center, Leiden, The Netherlands
    Search for more papers by this author
  • Eleonora P. M. Corssmit,

    1. Department of Endocrinology and Metabolism, Leiden University Medical Center, Leiden, The Netherlands
    Search for more papers by this author
  • Karel Pacak

    1. Reproductive and Adult Endocrinology Program, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD, USA
    Search for more papers by this author

Karel Pacak, Professor of Medicine, Chief, Section on Medical Neuroendocrinology, Reproductive and Adult Endocrinology Program, NICHD, NIH, Building 10, CRC, Room: 1E-3140, 10 Center Drive MSC-1109, Bethesda, 20892-1109 MD, USA. Tel.: +1 301-402-4594; Fax: +1 301-402-0884; E-mail: karel@mail.nih.gov

Summary

Formerly used concepts for phaeochromocytomas and paragangliomas have been challenged by recent discoveries that at least 24% of tumours are familial and thereby often multiple in various locations throughout the body. Furthermore, tumours are often malignant and perhaps more aggressive if associated with SDHB gene mutations. Some paragangliomas are clinically silent and may present only with dopamine hypersecretion. In the current era where CT and MRI are more commonly used, tumours are more often found as incidentalomas and MRI may be less specific for phaeochromocytoma and paraganglioma than previously thought. Because of unique tumour characteristics (e.g. the presence of cell membrane and intracellular vesicular norepinephrine transporters) these tumours were ‘born’ to be imaged by means of specific functional imaging approaches. Moreover, additional recent discoveries related to apoptosis, hypoxia, acidosis, anaerobic glycolysis and angiogenesis, often disturbed in tumour cells, open new options and challenges to specifically image phaeochromocytomas and paragangliomas and possibly link those results to their pathophysiology, genotypic alterations and metastatic potential. Functional imaging, especially represented by positron emission tomography (PET), offers an excellent approach by which tumour-specific processes can be detected, evaluated and seen in the context of tumour-specific behaviour and its genetic signature. In this review, we address the recent developments in new functional imaging modalities for phaeochromocytoma and paraganglioma and provide the reader with suggested imaging approaches in various phaeochromocytomas and paragangliomas of sympathetic origin. Current imaging algorithms of head and neck parasympathetic paragangliomas are not discussed. Finally, this review outlines some future perspectives of functional imaging of these tumours.

Introduction

International nomenclature and guidelines concerning phaeochromocytoma and paraganglioma, tumours that are derived from chromaffin tissues, have changed drastically over the last few years.1 Officially, the term phaeochromocytoma must be reserved for those paragangliomas located inside the adrenal glands, whereas sympathetic paragangliomas outside the adrenals are referred to as extra-adrenal paragangliomas. Most of these sympathetic paragangliomas are able to produce, metabolize and secrete catecholamines. Paragangliomas located in the head and neck region are often derived from parasympathetic tissue and rarely secrete catecholamines. Although the differentiation between intra- and extra-adrenal localization seems arbitrary, it reflects the distinctive biochemical and clinical properties of these tumours. Adrenal tumours are usually benign, secrete both epinephrine and norepinephrine in at least 50% of cases, are often related to a specific gene mutation if located bilaterally, and are frequently found as incidentalomas. In contrast, extra-adrenal tumours have a noradrenergic and/or dopaminergic phenotype and more often have an aggressive or metastatic nature. However, reports of clinically silent tumours that do not secrete catecholamines at all or merely dopamine are emerging.2 Recent studies have shown that at least 24% of paragangliomas are familial and thereby often multiple in various locations throughout the body.3,4 Because there is no definite histological substrate for malignant phaeochromocytoma, malignant disease can only be established by either demonstrating local tumour invasion and/or the presence of paraganglioma cells outside the normal sites. Interestingly, phaeochromocytomas and paragangliomas that are metastatic are most often related to mutations in the SDHB gene, further emphasizing the need to consider genotype–phenotype associations.5 Although sensitivity of MRI imaging in paraganglioma and phaeochromocytoma is high, its specificity may be disappointing and one study found that the ‘classical’ image of phaeochromocytoma was present in only a minority of patients.6 Therefore, functional imaging may be a good asset in imaging strategies in phaeochromocytoma and paraganglioma, especially with equivocal MRI results and high clinical and biochemical suspicion for the presence of a tumour.

As phaeochromocytoma and paraganglioma have specific cellular and intracellular characteristics, they may well be described as being born to be imaged specifically and uniquely (Fig. 1). These characteristics favour the use of functional imaging modalities, including situations when proof that a tumour is an adrenal phaeochromocytoma is needed, which is the case in: (i) norepinephrine secreting tumours (these tumours may also be located extra-adrenally), (ii) search for metastatic disease (especially in those tumours over 5 cm in size), (iii) familial phaeochromocytoma, especially due to their multiplicity or perhaps higher metastatic potential if associated with an SDHB mutation. Current functional imaging of endocrine tumours has revealed that these modalities, except for their well-known specific detection and localization of a tumour, have potential to assess: (i) tumour behaviour,7 (ii) response to therapy (e.g. the presence of specific transporters or receptors, apoptosis),7,8 (iii) genetic background (e.g. detection of oxidative stress)7,9 and (iv) the potential to metastasize (e.g. evaluation of angiogenesis).7,10,11 Recent introduction of combined PET/CT scans further increased precise detection and localization of tumours, which could eventually reduce cumulative cost for additional and multiple imaging modalities.12,13

Figure 1.

 Nonspecific and specific imaging for pheochromocytoma/paraganglioma. DBH, Dopamine-beta-hydroxylase; EPI, Epinephrine; 18F-FDA, 18F-fluorodopamine; 18F-FDG, 18F-fluoro-2-deoxy-d-glucose; 18F-FDOPA, 18F-dihydroxyphenylalanine; 18F-FNE, 18F-Free Norepinephrine; NE, Norepinephrine; DOTANOC, DOTA-Nal3-octreotide; DOTATOC, DOTA-Tyr3-octreotide; 123/131I-MIBG, 123/131I-metaiodobenzylguanidine; l-AADC, l-aromatic-aminoacid decarboxylase; NET, Norepinephrine transporter; PNMT, Phenylethanolamine-N-methyltransferase; SDHB/C/D, Succinate dehydrogenase subunit B/C/D; ST Receptor, Somatostatin receptor; ?*, Potential radiopharmaceutical directed at mutations in the mitochondria.

Established functional imaging of phaeochromocytoma and paraganglioma

Similar to the sympathetic nervous system, phaeochromocytomas and most extra-adrenal paragangliomas express cell membrane norepinephrine transporters (NET) through which catecholamines can enter cells to be stored in vesicles (Fig. 1). For many years metaiodobenzylguanidine (MIBG) has been used for diagnostic imaging in phaeochromocytoma because of its resemblance to norepinephrine and its uptake by the NET.14 [131I]-MIBG scintigraphy has a sensitivity of 77–90% and a specificity of 95–100%.15–17 In 1986, Shulkin et al.18 illustrated the superiority of scintigraphy with [123I]-MIBG over [131I]-MIBG in a paraganglioma patient. Later studies have shown that the use of the [123I]-isotope resulted in a better performance with a sensitivity of 83–100% and a specificity of 95–100%.17,19–23 In addition, the [123I]-isotope can be visualized with single photon emission computed tomography (SPECT) imaging, further increasing its diagnostic accuracy. Importantly, the normal adrenal medullary may show physiological uptake of both [131I]- and [123I]-MIBG.24,25 Suboptimal sensitivity of MIBG scintigraphy might be associated with the relatively lower affinity of MIBG to the NET in comparison to newer compounds, the lack of storage granules or the loss of transporters by tumour cell dedifferentiation.26 Furthermore, medication use interfering with MIBG uptake, e.g. calcium channel blockers, labetalol and tricyclic antidepressants, could result in false-negative results.27

Somatostatin receptors have also been discovered on paragangliomas and have been used in imaging of phaeochromocytomas and paragangliomas (Fig. 1). Although [111In]-pentetreotide (Octreoscan) can be used in phaeochromocytoma diagnosis, it has in general been considered inferior to the use of MIBG in patients with benign intra-adrenal phaeochromocytomas.28–34 In 2001, van der Harst et al.35 reviewed their experience of preoperative [123I]-MIBG scintigraphy and imaging with a labelled somatostatin analogue in the diagnostic work-up in phaeochromocytoma patients. The detection rate of the Octreoscan was only 25% for primary benign phaeochromocytomas, but metastases were detected in 7 of 8 cases. Thus, somatostatin receptor imaging might be considered as a supplement for MIBG scintigraphy in phaeochromocytoma and paraganglioma patients with suspected metastatic disease.

Emerging endocrine functional imaging of phaeochromocytoma and paraganglioma

The introduction of positron emission tomography (PET) has had a tremendous impact on functional imaging. [11C]-hydroxyephedrine ([11C]-HED) is more polar than MIBG, has even greater similarities with norepinephrine and was the first positron-emitting probe of the sympathoadrenal system used in humans.36,37 [11C]-HED synthesis is complex and it has a very short half-life of only 20 min, and thus requires onsite production for each patient. Newer positron-emitting compounds used in PET imaging, like [18F] (half-life 110 min), generate positrons that result in high resolution images. The advantage of [18F] is that the incorporation into a molecule has only small effects on the ability of the carrying compound to bind to receptors, or be taken up by their transporters.38 Their relatively short half-lives in comparison with MIBG increase the maximal dose that can be administered safely and make imaging possible shortly after injection, instead of the mandatory postponed imaging 24 and 48 h after MIBG injection. Furthermore, recent revolutionary developments enabled combining conventional imaging methods like CT with PET, further improving its diagnostic use.39 The positron-emitting compound [18F] may be used in combination with several carrier compounds relevant in phaeochromocytoma/paraganglioma. The compound dopamine is a much better substrate for the NET than norepinephrine itself.26 Clinical studies have confirmed that [18F]-fluorodopamine ([18F]-FDA) can be used as an imaging agent for phaeochromocytoma.40–43

In addition to expressing NET, phaeochromocytomas are neuroendocrine tumours and are able to take up and decarboxylate amino acids like dihydroxyphenylalanine (DOPA). DOPA can be labelled with [18F] to form the imaging compound [18F]-dihydroxyphenylalanine ([18F]-FDOPA). DOPA can be decarboxylated to dopamine by l-amino acid decarboxylase (l-AADC), which is shown in Fig. 1. At least part of this compound’s ability to localize paragangliomas is the fact that [18F]-FDOPA is converted to [18F]-FDA which is subsequently stored in the intracellular vesicles (Fig. 1). Carbidopa has been reported to enhance the sensitivity of [18F]-FDOPA PET for paragangliomas by further increasing the tumour-to-background ratio of tracer uptake.44

The most frequently used PET imaging agent is [18F]-fluoro-2-deoxy-D-glucose ([18F]-FDG). Imaging with [18F]-FDG PET reflects excessive glucose uptake mainly via GLUT-1 in metabolically hyperactive tumours (increased anaerobic glycolysis).45 Although [18F]-FDG PET is less specific, studies have shown it can be useful in patients with phaeochromocytoma and paraganglioma, especially in those with malignant disease, as will be discussed below.46,47

Newer somatostatin analogues like DOTA-Tyr3-octreotide (DOTATOC) have shown favourable characteristics in imaging with high affinity for somatostatin receptors and a stable and easy process of labelling.48 Although most existing somatostatin-based tracers only have affinity for the somatostatin receptor subtype 2, which is not always present on phaeochromocytoma and paraganglioma cells, newer compounds, such as DOTA-Nal3-octreotide (DOTANOC), also have affinity for other somatostatin receptor subtypes.49,50 At present, these new derivatives like DOTATOC and DOTANOC labelled with the PET radiotracer [68Ga] have shown promising results in imaging of somatostatin receptor positive tumours as compared to the non-PET [111In]-pentetreotide scintigraphy, exemplifying the superior performance of PET imaging over scintigraphy in general.49,51,52

The need for an individualized approach in the functional imaging of phaeochromocytomas and paragangliomas

Data published about phaeochromocytoma and paraganglioma support the notion that the following contemplations must be taken into account in imaging approaches to these tumours in each patient: the suspicion for adrenal or extra-adrenal disease, the risk of recurrent metastatic or multifocal disease, and the suspicion for familial tumours, especially the presence of succinate dehydrogenase (SDHx) gene mutations. Previous and recent articles characterizing very well described phenotypes of these tumours must guide the use of proper imaging modalities in close context with their clinical and biochemical phenotypes.53–55 The biochemical phenotype may also point towards preferred initial genetic testing that further supports the choice of initial imaging approach and algorithm.

Below we will discuss the evidence of implementing a tailor-made functional imaging approach in these different subsets of patients with phaeochromocytoma and paraganglioma. Functional imaging in adrenal vs. extra-adrenal disease will be reviewed in sections ‘Functional imaging approach to adrenal paragangliomas (phaeochromocytomas)’ and ‘Functional imaging approach to extra-adrenal paragangliomas’, functional imaging in metastatic disease will be addressed in section ‘Functional imaging approach to metastatic phaeochromocytomas and paragangliomas’ and the presence of SDH mutations and their putative role in the performance of functional imaging modalities in phaeochromocytomas and paragangliomas is discussed in section ‘Suspicion of the presence of SDH mutations and their putative role in the performance of functional imaging modalities in phaeochromocytomas and paragangliomas’.

Adrenal vs. extra-adrenal disease

Functional imaging approach to adrenal paragangliomas (phaeochromocytomas)

About 80% of the sympathetic paragangliomas arise from the adrenal medulla and are thus ‘true’ phaeochromocytomas.4 Therefore, some reports have questioned the need for routine use of functional imaging in the diagnostic workup for phaeochromocytoma, suggesting that the treatment plan in patients with an adrenal lesion suspicious for phaeochromocytoma, in the absence of hereditary disease or a history of phaeochromocytoma, will not change.56,57 It was concluded that in nonfamilial cases with a clear biochemical diagnosis and a unilateral adrenal mass on CT or MRI, no additional functional imaging is necessary.57,58 Although this recommendation may be valid in certain cases, the following issues need to be addressed. Determination of the biochemical phenotype must be taken into account in the decision making. As is shown in Fig. 2, epinephrine is the end product after norepinephrine is changed by phenylethanolamine N-methyltransferase, an enzyme that is expressed in the adrenal glands but not in extra-adrenal lesions (except extremely rare cases). Therefore, hypersecretion of only epinephrine and its respective metanephrine metabolite (the adrenergic phenotype) reflects the presence of an adrenal mass, whereas patients that have a noradrenergic biochemical phenotype may have either an adrenal, an extra-adrenal lesion, or both.

Figure 2.

 Pathways of metabolism of catecholamines to free and sulphate-conjugated metanephrines. Courtesy of: Dr G. Eisenhofer. PNMT, phenylethanolamine N-methyltransferase; COMT, catechol-O-methyltransferase; SULT1A3, monoamine-preferring sulfotransferase.

Furthermore, research has shown that apparently sporadic tumours were found to be hereditary in at least 24% of cases, which could lead to an underestimated risk for multifocal and malignant disease and indicate that absence of a family history may not be used as a criterion for not performing functional imaging.3,59 Paragangliomas in patients with multiple endocrine neoplasia type 2 (MEN2) are usually located in the adrenal and always produce epinephrine and/or metanephrine, alone or together with norepinephrine and/or normetanephrine. Interestingly, patients with MEN2 are not only at risk for developing bilateral adrenal disease but may also have multiple tumours in one adrenal gland. On the other hand, lack of phenylethanolamine-N-methyltransferase activity results in a solely noradrenergic phenotype in von Hippel-Lindau syndrome (VHL) associated adrenal tumours. Familial paragangliomas, those related to SDHx gene mutations, have been associated with tumours in adrenal but more often extra-adrenal locations and may present with multifocal or metastasized disease.

Even though MRI and CT have excellent sensitivity (90–100%),40 the ability of CT and MRI alone to specify a phaeochromocytoma from other abdominal lesions is insufficient.20 Recently, Jacques et al.6 demonstrated the wide range of possible appearances of phaeochromocytoma on MRI images, emphasizing their low specificity and questioning the relevance of finding the ‘classical’ hyperintense phaeochromocytoma image on a T2 weighted MRI. Thus, the current concept of any area-limited imaging may be insufficient in these patients in whom multifocal, extra-adrenal and even metastatic disease cannot be easily ruled out.

Several reports have shown [123I]-MIBG scintigraphy to have a decent performance in intra-adrenal phaeochromocytomas with sensitivities ranging from 85% to 100%. However, a recent study by Bhatia et al.60 in this journal eloquently displayed the likelihood of false-negative MIBG scintigraphies in patients with smaller, albeit mainly sporadic, tumours. Importantly, because screening in asymptomatic stages of disease in patients with a hereditary risk is increasingly advocated, we expect these smaller tumours to be encountered more frequently in the future.60 Physiologic uptake of [123I]-MIBG and [18F]-FDA in normal adrenal glands may lead to false-positive results. The use of standardized uptake values for distinguishing adrenal glands with phaeochromocytoma from those without, has been advocated in [18F]-FDA PET.61 Phaeochromocytoma is unlikely with SUV below 7·3, whereas an SUV above 10·1 confirms the presence of phaeochromocytoma. Imaging with [18F]-FDOPA PET outperformed [123I]-MIBG scintigraphy in the detection of phaeochromocytoma.62 In addition, to its advantage, it was noted that normal adrenal glands lack uptake of [18F]-FDOPA,63,64 whereas both the [123I]-MIBG and [18F]-FDA compounds revealed some degree of accumulation in normal adrenal glands.

Newer studies will have to address specifically the performance of functional imaging modalities in relation to the underlying gene mutation. Although data concerning functional imaging in patients specifically with a MEN2 associated phaeochromocytoma is scarce, [123I]-MIBG scintigraphy, [18F]-FDA PET and [18F]-DOPA PET are thought to perform well, with the advantages of the newer PET imaging characteristics as described above. In contrast, in patients with VHL-associated phaeochromocytoma, [18F]-FDA PET significantly outperformed [123I]-MIBG scintigraphy, which could be related to the limited expression of NET in VHL paraganglioma cells.26,43 The fact that [18F]-FDA has a much higher affinity for these receptors than MIBG is thought to be responsible.43,65

In conclusion, patients with an adrenal lesion suspect for phaeochromocytoma need imaging with either [123I]-MIBG scintigraphy, [18F]-FDA PET or [18F]-DOPA PET to detect or exclude multifocal or metastatic disease (Table 1). However, in the case of a noradrenergic or dopaminergic phenotype, additional paragangliomas located outside the adrenals need to be excluded. Albeit more expensive, we believe imaging quality and favourable dosimetry favour the use of the newer PET imaging compounds, where available. Furthermore, since a hereditary basis is found to be much more frequent than previously estimated, genetic results must certainly be taken into account. Adrenal tumours associated with a VHL gene mutation, are best imaged by [18F]-FDA PET, whereas studies specifically concerning adrenal paragangliomas associated with other genotypes (i.e. MEN) are awaited.

Table 1.   Suggested functional imaging approaches according to clinical phenotype
 Relevant text sectionSuggested modalities  
  1. PGLs, paragangliomas. *If available, PET imaging is preferred because of superior sensitivity and dosimetry, †Adrenal PGLs in von Hippel-Lindau syndrome are better visualized by [18F]-FDA PET in comparison to MIBG, ‡[18F]-FDOPA PET may perform poorly in SDHB associated metastatic disease. §[18F]-FDG PET superior in SDHB associated metastatic disease, ¶If available, newer somatostatin analogues preferred over pentetreotide. For references: see relevant text sections.

Adrenal PGLs (phaeochromocytomas)Functional imaging approach to adrenal paragangliomas (phaeochromocytomas)[18F]-FDA PET*†[18F]-FDOPA PET*[123I]-MIBG scintigraphy
Extra-adrenal PGLsFunctional imaging approach to extra-adrenal paragangliomas[18F]-FDA PET[18F]-FDOPA PET‡ 
Metastatic PGLsFunctional imaging approach to metastatic phaeochromocytomas and paragangliomas[18F]-FDA PET[18F]-FDG PET§[18F]-FDOPA PET‡
  [111In]-pentetreotide scintigraphy¶[68Ga]-DOTATOC/DOTANOC PET*¶[123I]-MIBG scintigraphy (assess eligibility for [131I]-MIBG treatment)

Functional imaging approach to extra-adrenal paragangliomas

Almost all extra-adrenal paragangliomas lack the phenylethanolamine-N-methyltransferase (Fig. 2) and, therefore, produce norepinephrine and its metabolite normetanephrine without elevated levels of epinephrine and its derivative. Extra-adrenal locations of paragangliomas are frequently found in familial paraganglioma syndromes associated with the SDHx mutations.53–55 By contrast, extra-adrenal and malignant disease are rare in MEN2 patients. Malignant disease is most prevalent in SDHB carriers (up to 70%), whereas metastases are thought to be very rare in SDHD mutation carriers (∼2·5%).66,67 As patients have a high risk for multifocal and/or metastatic disease and a negative family history is not uncommon, the need to perform whole body functional imaging must be emphasized in these patients.

Extra-adrenal paragangliomas have been frequently reported to lack adequate uptake of [123/131I]-MIBG.35,60,68 In our experience, extra-adrenal lesions may be easily missed on CT or MRI if the suspicion of their presence is not specifically indicated. Overall, [18F]-FDA PET has a better performance in detecting extra-adrenal paragangliomas than [123I]-MIBG scintigraphy and should therefore be the preferred imaging modality, if available. In a study of 17 patients with nonmetastatic adrenal and extra-adrenal paraganglioma, [18F]-DOPA PET detected tumours with a strikingly high sensitivity and specificity of 100% for both.62 Interestingly, our study of patients with predominantly metastatic paraganglioma revealed a sensitivity for [18F]-DOPA PET of only 50%.44 However, these results may have reflected an overrepresentation of patients with SDHB mutations and their associated high prevalence of malignant disease, which will be further discussed below.

In conclusion, patients may present with extra-adrenal paragangliomas and clinicians need to realize that the accuracy of imaging with MIBG in these extra-adrenal tumours is often disappointing. Functional PET imaging with [18F]-FDOPA or [18F]-FDA has been reported to be a better approach (Table 1). Interestingly, in patients with metastasized SDHB associated paragangliomas, the [18F]-FDOPA PET may perform very poorly and should therefore not be advised.

Functional imaging approach to metastatic phaeochromocytomas and paragangliomas

In the absence of specific nuclear characteristics in malignant phaeochromocytoma and paraganglioma, the histopathological diagnosis of malignancy is impossible and the diagnosis of metastatic disease is based on the findings of tumour tissue at locations where chromaffin cells are normally not present. Metastatic phaeochromocytomas and paragangliomas are most commonly associated with mutations in the SDHB gene, in an estimated 40% of cases if primary tumours are located in the abdomen.5 Adrenal paragangliomas (phaeochromocytomas) associated with MEN and VHL are rarely metastatic. Although a heterogeneous appearance on MRI results in a higher degree of suspicion for malignancy, its predictive value is low.6 Metastatic paragangliomas imaged by [131/123I]-MIBG scintigraphy have often been found to be false negative or suboptimal.35,47,69 Therefore, currently, [131/123I]-MIBG scintigraphy in these patients is recommended to be performed only to evaluate whether the patient qualifies for [131I]-MIBG treatment, when other functional imaging modalities like e.g. [18F]-FDA PET are available. Labelled somatostatin analogues like [111In]-pentetreotide may be of additional value to [123I]-MIBG in the diagnostic work-up of patients with suspected metastatic paraganglioma and phaeochromocytoma.35 However, [111In]-pentetreotide scintigraphy lacks tissue specificity, only revealing the tumour’s somatostatin receptor status. Reports have indicated a higher sensitivity of somatostatin receptor based imaging in detecting MIBG negative metastatic disease and in dopamine-secreting tumours.34,70 Newer positron-emitting somatostatin analogues like [68Ga]-DOTATOC and DOTANOC reveal promising results.52 Importantly, both the expression of NET and somatostatin receptors may be lost in dedifferentiated tumours, resulting in false-negative imaging in metastatic disease.32,35,60

Superiority of [18F]-FDA PET over [131/123I]-MIBG scintigraphy was shown in malignant tumours.47,69 The sensitivity of [18F]-FDOPA PET for metastatic paragangliomas is reported to be limited, but preliminary results of an ongoing study show [18F]-FDOPA PET may perform especially poorly in patients with SDHB-related metastatic disease.44 As malignant tumours are generally metabolically more active, [18F]-FDG PET imaging is a very useful approach, albeit less specific. In a large study concerning SDHB-associated malignant paragangliomas, [18F]-FDG PET was found to be the most sensitive imaging method by far.47 In another study focusing on paraganglioma bone metastases specifically, bone scintigraphy proved useful in the staging of patients with malignant phaeochromocytoma and paraganglioma. As for other functional imaging, [18F]-FDG PET was highly recommended in SDHB mutation positive patients, whereas [18F]-FDA PET was recommended in patients without the mutation.71 With increasing dedifferentiation of the tumour, the ideal functional imaging modality appears to shift from the more specific [18F]-FDA PET to the less specific [18F]-FDG PET, the so called ‘flip-flop’ phenomenon.72 However, a larger study evaluating the role of [18F]-FDG PET in various phaeochromocytoma and paraganglioma is missing.

In conclusion, imaging with [123I]-MIBG scintigraphy has a limited role in metastatic phaeochromocytomas and paragangliomas, unless it is used to determine whether or not a patient is eligible for [131I]-MIBG treatment. Table 1 summarizes the suggested imaging modalities. We would prefer the use of [18F]-FDA PET imaging in patients with malignant phaeochromocytoma and paraganglioma without a known genetic mutation. The performance of [18F]-FDOPA PET may be limited in malignant phaeochromocytoma and paraganglioma, but this might be related to only those with an SDHB mutation. Somatostatin analogues like [111In]-pentetreotide, [68Ga]-DOTATOC and [68Ga]-DOTANOC may be of use in imaging, although somatostatin receptors may be lost during the process of dedifferentiation. In metastatic tumours associated with an SDHB mutation, [18F]-FDG PET has shown its superiority.47 Nonetheless, because a large proportion of malignant phaeochromocytomas and paragangliomas are associated with mutations in the SDHB gene per se, one might speculate that the excellent performance of [18F]-FDG PET in metastasized SDHB associated tumours, is a reflection of this genetic background itself.

Suspicion of the presence of SDH mutations and their putative role in the performance of functional imaging modalities in phaeochromocytomas and paragangliomas

Familial paraganglioma syndromes are associated with SDH gene mutations. SDH consists of four subunits (subunit A, B, C and D) of the mitochondrial complex II and is involved in two key mitochondrial pathways: the inner mitochondrial membrane bound electron transport chain and the mitochondrial matrix associated Krebs tri-carboxylic-acid (TCA) cycle. The interaction between both is necessary for maximal efficiency in ATP (energy) production under aerobic conditions. Mutations may lead to complete loss of SDH enzymatic activity in paragangliomas, with subsequent deregulation of hypoxia responsive genes.73 Impairment of mitochondrial function due to loss of SDH function may cause tumour cells to shift from oxidative phosphorylation to anaerobic glycolysis, also known as the ‘Warburg effect’.74 Therefore, we hypothesize that avid [18F]-FDG uptake by paragangliomas in SDHB positive patients does not merely reflect a high metabolic rate due to malignancy per se, but could be linked to SDHB-mutation specific tumour characteristics associated with upregulated glycolytic pathway. Although this has not been confirmed on a molecular level, other means of functional imaging may depend on the underlying mutations in a similar fashion. Future functional imaging approaches in phaeochromocytomas and paragangliomas could possibly provide clinicians with a link to their pathophysiology, genotypic alterations and metastatic potential. Functional imaging (especially PET) offers an excellent approach by which tumour specific processes can be detected, evaluated and seen in the context of tumour-specific behaviour and its genetic signature. Future studies will have to generate more data specifically focused on [18F]-FDA, [18F]-FDOPA and [18F]-FDG PET imaging performances in separate patient cohorts with different underlying mutations. Head-to-head comparisons of imaging modality performance by mutation are awaited.

Conclusions, recommendations and future prospects

Paragangliomas and phaeochromocytomas are tumours that are ideal for functional imaging because of their unique tumour characteristics. Whole body functional imaging may be a rational approach in most cases, definitely in those with large, extra-adrenal and SDHB-related tumours. Cost and availability of the PET compounds, however, still play a considerable role. From currently available PET radiopharmaceuticals, the use of [18F]-FDA or [18F]-FDOPA in patients with a biochemically established diagnosis of paraganglioma (often initially with unknown genotype) is warranted, especially when the aim is to localize the primary tumour and to rule out multifocality or metastases. Although [123I]-MIBG scintigraphy appears to perform reasonably well in primary benign adrenal paragangliomas (phaeochromocytomas), its sensitivity drops in extra-adrenal or malignant paragangliomas and it lacks the favourable dosimetry of the newer modalities.

Underlying gene mutations in phaeochromocytomas and paragangliomas are much more prevalent than previously expected and should be considered when choosing the appropriate functional imaging modality. For example, in VHL carriers [18F]-FDA PET has been found to be superior compared to [123I]-MIBG scintigraphy. Preliminary results of an ongoing study show that in patients with SDHB mutations [18F]-FDG or [18F]-FDA PET should be imaging modalities of choice, and [18F]-FDA or [18F]-DOPA PET are methods of choice in patients where SDHB has been ruled out. If PET imaging is not available, scintigraphy with [111In]-pentetreotide and/or the newer somatostatin analogues could be useful in addition to [123I]-MIBG, mainly in detection of metastatic disease. SDHB-related phaeochromocytomas and paragangliomas have a very high risk for the development of malignant disease and appear to have a distinctive ‘imaging phenotype’ with a shift towards higher uptake of the less specific [18F]-FDG compound compared to the more specific PET compounds (flip-flop). Whether this is a reflection of dedifferentiation and metastatic disease, or is directly associated with mutations in the SDH genes, is subject to further studies.

Newer functional imaging approaches may detect and evaluate tumour responses to therapy well before the actual decrease in tumour size in the context of its genetic signature and, thus, monitor the activity of the tumour itself. Therapeutic trials in the future will have to be done in close collaboration with PET departments to create an individually focused treatment approach. Of course, especially in those patients with the need for repetitive imaging because of hereditary risk, the possible risks of radiation from contrast CT and PET and from gadolinium in MRI should be taken into account in every patient. We believe efficient use of the right functional imaging method with an individualized rational approach may lead to reduced costs and better diagnostics. Future studies will further focus on unravelling pseudohypoxia, apoptosis and angiogenesis as potential targets for functional imaging in these tumours.

Acknowledgements

This research was supported, in part, by the Intramural Research Program of the NIH, NICHD. We would like to thank Dr H.J.L.M. Timmers for his helpful comments in the preparation of this manuscript.

Competing interests/financial disclosure

The authors have nothing to disclose.

Ancillary