Total-Body PET Imaging of Musculoskeletal Disorders

Introduction

Musculoskeletal disorders (MSDs) are broadly defined as diseases or injuries of the joints, bones, muscles, nerves, tendons, ligaments, supporting soft tissues, cartilage, and spinal discs, and comprise over 150 diagnoses.¹ Commonly encountered examples are arthritis (such as rheumatoid arthritis, systemic lupus erythematous, or osteoarthritis), osteoporosis/osteopenia, sarcopenia, infection, neoplasms, and neck and lower back pain. MSDs are a major cause of disability worldwide, second only to mental and substance use disorders,² and have a significant negative impact on global population health. They are prevalent across the lifespan and have an astronomical societal cost that comes in the form of limited mobility, dexterity and functional ability, persistent pain, inability to work, early retirement from work, inability to participate in societal roles, depression, and increased risk of developing other chronic disease conditions.³

Clinical evaluation of MSDs mostly relies on a combination of physical examination, serum biomarkers and pain- and health-assessment questionnaires. Outside of radiography, other imaging modalities have had a limited role in MSD evaluation. Anatomical imaging data from radiography, computed tomography (CT), and magnetic resonance imaging (MRI) have been shown to be only weakly correlated with clinical signs and disease pathogenesis or treatment response.⁴ Therefore, precise image-based parameters that could allow assessment of the molecular and pathophysiologic processes that underlie the disease state and its progression over time could play a crucial role in the evaluation of MSDs. Molecular imaging with positron emission tomography (PET)/CT and PET/MRI has been employed for MSD evaluation, but only in a limited number of studies.⁵ This has been a direct consequence of the associated high cumulative radiation dose, especially in the context of evaluating chronic MSDs, patient discomfort from the relatively long image acquisition time, and the inability to examine multiple body structures or systems that that are known to be affected by MSD. Furthermore, the limited spatial resolution of standard PET scanners is suboptimal for visualizing and quantifying disease activity in small joints or tissues, which appear to be a bellwether of pathology in several MSDs.

In this chapter we provide our perspective about the potential role of total-body PET/CT (TB PET/CT) in the management of patients with MSDs, with emphasis on autoimmune and degenerative arthritis, musculoskeletal infection, osseous disorders, sarcopenia, and musculoskeletal malignancies. Throughout this scientific communication, we elaborate on three major physical characteristics of TB PET/CT systems that are deemed crucial for successful evaluation of MSD disorders: (1) the long axial field-of-view, which enables imaging multiple organs and systems simultaneously during the same phases of radiotracer distribution and uptake; (2) high geometric sensitivity, which allows the reduction of the administered radiotracer dose significantly and the shortening of the image acquisition time; and (3) higher spatial resolution compared to standard PET systems, which is essential for detecting and quantifying disease process in the small musculoskeletal structures such as small joints, tendons, and soft tissue structures.

Rheumatoid and Psoriatic Arthritis

Rheumatoid arthritis (RA) and psoriatic arthritis (PsA) are two common autoimmune disorders that are systemic in nature and cause an inflammatory reaction, particularly in the joints. They are a major cause of disability and loss of function in the adult population worldwide. The underlying causes of these disorders are unknown, and the majority of patients have no family history of these serious musculoskeletal disabilities. While the prevalence of both conditions increases by age, juvenile idiopathic arthritis (JIA), formerly known as juvenile rheumatoid arthritis (JRA), is now recognized as an independent clinical entity in the broad class of autoimmune arthritis.

Joint inflammation is considered the hallmark of autoimmune arthritis, and tenosynovitis (inflammation of the synovium and the tendons) is commonly presented in both RA and PsA. The entity named enthesitis (inflammation of the entheses), which is unique to PsA, is considered the hallmark for early tissue damage. The existing treatment modalities include a vast number of established approaches and new therapeutics, such as tumor necrosis factor inhibitors, that target joint inflammation^6,7 and broadly show successful modification of the disease course.^8–10 Clinical evaluation of RA and PsA (based on a physical exam performed by a rheumatologist and serum biomarkers) underestimates disease burden and has limited sensitivity and specificity in assessing response to therapeutic interventions.^11,12 Conventional radiographs, MRI, and ultrasound scanning, while useful for assessing structural consequences of these disorders, do not directly determine disease activity at the cellular level, which is essential for early detection and determining response to therapy.¹³ By now it is well established that structural abnormalities are a late manifestation of the disease process. As such, discrepancy and discordance between structural imaging findings in the evaluation of inflammatory arthropathies are not uncommon.¹⁴ Recent advancement in quantitative and semi-quantitative standard anatomical imaging, such as contrast-enhanced ultrasound,^15,16 quantitative MRI,¹⁷ and dual-energy CT iodine maps¹⁸ have been somewhat successful in the evaluation and staging of inflammatory arthropathies and their response to treatment. However, further studies are necessary to confirm their accuracy and practicality in this setting. PET/CT imaging with radiotracers such as ¹⁸F-fluorodeoxyglucose (FDG) and ¹⁸F-sodium fluoride (NaF) has been shown to be highly sensitive and accurate in examining MSDs, particularly those with significant inflammation (Figure 1).¹³

Figure 1.

TB PET in rheumatoid and psoriatic arthritis. Maximum intensity projection (MIP) from the TB PET scan of a 65-year-old man with established rheumatoid arthritis (A). Images of the subject in (A) showing classic ring-like patterns of radiotracer uptake consistent with synovitis in the joints of the hand (B), and foot (C). Extensor and flexor tenosynovitis and enthesitis in the right second digit of a 72-year-old man with established psoriatic arthritis (D), also shown in the MIP of the TB PET scan (E). These images were acquired on the uEXPLORER PET/CT scanner at the University of California Davis, with an injected dose of about 74 MBq. Images show static scans conducted over 20 min, starting at 40 min post-radiotracer injection.

Courtesy of Y. Abdelhafez, MD, University of California Davis

FDG-based PET/CT imaging has been frequently employed for assessing RA and PsA at various stages of the disease and monitoring response to treatment. ^13,19–28 In one research study by Raynor et al., global synovial FDG activity was measured in 19 RA patients.²⁹ These investigators employed a thresholding algorithm (ROVER software, ABX GmbH, Radeberg, Germany) to delineate focal FDG uptake in the synovial joints in the hands, elbows, shoulders, knees, and feet. PET parameters reflecting volume and uptake values with and without partial volume correction were found to be significantly higher in RA patients compared to healthy control subjects and correlated significantly with clinical features such as CD reactive protein, erythrocyte sedimentation rate, and swollen joint count. In addition to the joints assessed by Raynor et al., FDG-PET/CT has been proposed to assess involvement of the hip and temporomandibular joint in patients with RA.^30,31 A study by Jonnakuti et al. used PET imaging with NaF, a tracer that portrays osteoblastic activity, to assess knee involvement in RA.³² The authors noted that increased NaF uptake was associated with increased knee degeneration determined by Kellgren-Lawrence grading; therefore, they concluded that NaF-PET/CT may have clinical utility in assessing bone changes in RA. These studies emphasize the utility of TB PET/CT in facilitating the assessment of joints throughout the body, which are frequently involved in systemic disorders such as RA and PsA and can be effectively evaluated with this technology (Figures 2 and 3).

Figure 2.

FDG-PET maximum intensity projection (MIP) of a 69-year-old man with rheumatoid arthritis showing the upper body (A) and lower body (B). Synovitis was assessed by segmenting FDG-avid joints using an adaptive thresholding algorithm (ROVER software, ABX GmbH, Radeberg, Germany). Metabolically active volume (MAV), maximum standardized uptake value (SUV_max), mean SUV (SUV_mean), partial volume-corrected SUV_mean (pvcSUV_mean), total lesion glycolysis (TLG = MAV × SUV_mean), and partial volume-corrected TLG (pvcTLG = MAV × pvcSUV_mean) were calculated and summed for each segmented region. The global pvcTLG for this patient was 820.0. These analyses could be enabled at a lower radiation dose and scan time, with more comprehensive body coverage in a single scan, using TB PET/CT.

From Saboury B, Morris MA, Nikpanah M, Werner TJ, Jones EC, Alavi A. Reinventing Molecular Imaging with Total-Body PET, Part II: Clinical Applications. PET Clin. 2020;15(4):463-475. doi:10.1016/j.cpet.2020.06.013; with permission.

Figure 3.

NaF-PET maximum intensity projection (MIP) of the same rheumatoid arthritis patient as Figure 2 showing the upper body (A) and lower body (B). ROVER software was used to segment focal areas of high bone formation in the joints. Metabolically active volume (MAV), maximum standardized uptake value (SUV_max), mean SUV (SUV_mean), partial volume-corrected SUV_mean (pvcSUV_mean), total calcium metabolism (TCM = MAV × SUV_mean), and partial volume-corrected TCM (pvcTCM = MAV × pvcSUV_mean) were calculated and summed for each segmented region. The global pvcTCM for this patient was 1898.8.

From Saboury B, Morris MA, Nikpanah M, Werner TJ, Jones EC, Alavi A. Reinventing Molecular Imaging with Total-Body PET, Part II: Clinical Applications. PET Clin. 2020;15(4):463-475. doi:10.1016/j.cpet.2020.06.013

Several concerns have been raised about the role of PET/CT in the autoimmune arthritis population,³³ such as significant ionizing radiation exposure during the course of the disease and monitoring of treatment response, long scan times in a population that is functionally impaired, limited spatial resolution, which may be suboptimal for evaluating small joints such as those of the extremities that are affected early in the disease course, and examination of only a small anatomic segment of the body despite the clear need for systemic assessment. As noted above, the unique characteristics of TB PET/CT including the long axial field-of-view, high sensitivity, and excellent spatial resolution could overcome the challenges that are faced with conventional PET instruments. Particularly, TB PET/CT would enable (i) staging of RA or PsA disease activity in the entire body, as opposed to examining only a limited subset of joints, so that the appropriate therapy for each individual patient and disease state can be adopted and therefore improving the outcome;^34,35 (ii) monitoring response to therapy in joint tissues to optimize the therapeutic window³⁶ and drug efficacy on a personalized basis,^37–39 switching promptly to another treatment in non-responders;^40,41 and (iii) assessing the influence of treatment on other crucial organ system, such as the cardiovascular system,⁴² to reduce co-morbidities and side-effects.^43–45 In the juvenile population, the potential of lowering radiation dose and shortening scan time will also be of great importance since the incidence of autoimmune arthritis is relatively high in this age group.

Based on published literature, the majority of research studies in autoimmune arthritis have employed FDG as the main PET tracer. Molecular and cellular events that underlie the inflammatory-proliferative cascade in autoimmune arthritis include activation and transmigration of leucocytes, aberrant pathways of T-cell activation, angiogenesis, hypoxia, and increased osteoclastic and impaired osteoblastic activity and appear to play an important role in the pathogenesis of these disorders.^46–48 There are significant gaps in the knowledge regarding the exact role of these biological processes and their interactions during autoimmune arthritis pathogenesis. With the growing number of PET ligands,⁴⁹ these processes could be effectively examined by TB PET/CT to determine their role in these serious and yet treatable systemic inflammatory disorders. Such targeted research studies can also be of value in developing effective drugs in the future. While most PET imaging studies are performed as a single static scan at a late time point (mostly at 1-2 hours after the administration of the tracer), performing dynamic imaging with TB PET/CT may lead to improved knowledge in these conditions.^50,51

In summary, employing TB PET/CT imaging will lead to a significantly lower radiation dose, shorter scan time, and optimal quantification of the disease burden of autoimmune arthritides such as RA and PsA. This could address significant obstacles and longstanding challenges that have been faced in the fields of rheumatology about staging disease activity, treatment options, long-term monitoring, and assessing treatment response.

Osteoarthritis and Degenerative Conditions

Osteoarthritis (OA) is a degenerative disease that is frequently associated with aging, obesity, and prior joint trauma or secondary to inflammatory arthropathies (such as RA and gout) or avascular necrosis.⁵² This disorder represents one of the leading causes of impaired mobility in the elderly, and its prevalence is expected to increase as global life expectancy increases in the coming decades.⁵³ Although OA most commonly affects the spine, hip, knee, hands, and feet, it can potentially involve any synovial joint.⁵⁴ OA is characterized by systemic joint inflammation, exacerbated by proinflammatory changes that are related to aging as well as obesity,^55,56 and therefore this disease is an excellent candidate for imaging by PET/CT and PET/MRI.^57,58 In particular, FDG-PET/CT, as a validated and effective method for detecting and quantifying synovitis due to OA, is ideally suited for assessing this very common joint disease.⁵⁷ Wandler et al. found that uptake of FDG in the shoulder was associated with OA, bursitis, frozen shoulder, and rotator cuff injury.⁵⁹ FDG uptake in the knee synovium of symptomatic OA patients has been found to be higher compared to that of control subjects.⁶⁰ In addition, FDG uptake in the knee has been associated with aging as well as clinical symptoms of OA unrelated to aging.^61,62

In a recent study by Al-Zaghal et al., the role of FDG to assess degenerative changes in the knee was compared to that of NaF.⁶³ CT segmentation was used to quantify uptake of FDG, as an inflammatory biomarker, and NaF, as a tracer that reflects subchondral bone formation and turnover. Soft tissue FDG uptake was correlated with aging, and uptake of both tracers was correlated with patient body mass index. In a study of 34 subjects, Khaw et al. used NaF-PET/CT to evaluate bone turnover in the elbows, knees, hands, and feet.⁶⁴ PET segmentation using ROVER software facilitated the quantification of the volumetric and metabolic parameters of NaF-avid regions, which were used to determine global disease activity (Figure 4). Global PET parameters were found to correlate with subject body weight, suggesting that NaF-PET/CT is a sensitive method of determining the effects of biomechanical insufficiency on joint degeneration. To compare PET findings with those on MRI, Savic et al. used NaF-PET/MRI to evaluate changes related to OA in 16 patients with varying levels of disease.⁶⁵ The authors found that bone turnover determined by NaF-PET was correlated with degenerative changes in cartilage assessed by MRI. In two separate studies examining the hip joint, Kobayashi et al. demonstrated that NaF-PET can detect OA-related changes earlier than MRI and radiography and that NaF uptake correlated with pain severity.^66,67 Other studies have suggested that NaF uptake in the knee can identify changes before manifestation of abnormal findings on MRI.^68,69

Figure 4.

NaF-PET maximum intensity projection (MIP) of a 72-year-old man showing tracer uptake in the joints before (A) and after (B) PET segmentation using an adaptive thresholding algorithm (ROVER software). Volumetric and metabolic parameters were automatically calculated and summed to determine total disease activity. The global partial volume-corrected total calcium metabolism for this patient was 338.6.

In addition to OA involving the appendicular skeleton, degenerative processes affecting the spine such as disc degeneration and spondylosis also cause inflammatory reactions and increased bone turnover that are detected as abnormal sites on PET. In a study of 43 healthy subjects, FDG uptake in the thoracic and lumbar spine was found to correlate with body weight, likely representing early inflammatory changes related to degeneration.⁷⁰ A similar study demonstrated an association between body weight and increased NaF uptake in the cervical, thoracic, and lumbar spine.⁷¹ Rosen et al., who analyzed FDG-PET/CT images from 150 subjects, found that degenerative disk and facet disease present on CT correlated with FDG uptake.⁷² PET imaging can also be used to diagnosis of other sources of back and neck pain that often cannot be visualized by other modalities and include radiculopathy, spondylitis, spondylodiscitis, and postsurgical complications and infections.^73–81 The findings from these studies suggest that PET has a potential role is assessing unknown causes of back pain in addition to detecting degenerative changes.

Since diseases affecting the skeleton can occur throughout the body, TB PET/CT is uniquely situated to diagnose and assess these common disorders. With the data available from TB PET/CT, future studies will be able to assess the total disease burden in the entire body rather than focusing on a subset of skeletal sites. FDG and NaF have been demonstrated to be feasible markers of degenerative processes, and global assessment of uptake of these tracers may have a future role in the determination of disease severity and response to therapy. PET tracers beyond FDG and NaF that target other pathologic processes involved in joint degeneration, such as angiogenesis or macrophage activity, potentially may reveal novel therapeutic targets.

Fever of Unknown Origin

The classic definition of fever of unknown origin (FUO) was first described in 1961 by Petersdorf and Beeson, who defined it as: “fever higher than 38.3°C (100.9°F) on several occasions, persisting without diagnosis for at least 3 weeks in spite of at least 1 week’s investigation in hospital.”^82,83 The definition of FUO was later revised and now includes cases in which three outpatient visits have not resulted in a diagnosis.⁸⁴ Causes of FUO include infection, inflammatory diseases, and malignancy, with more than 200 differential diagnoses being recognized.^84–86 Musculoskeletal infections are an important cause of FUO. Although conventional modalities, such as radiography, ultrasound, CT, and MRI are typically first-line options, by definition, in a majority of these cases no known musculoskeletal source of infection can be suspected and imaged specifically by these modalities. Therefore, a total-body imaging approach is of paramount importance and advantage in this setting. Moreover, equivocal findings on structural imaging modalities may warrant further evaluation with molecular modalities. In recent years, imaging with FDG-PET has become the study of choice in patients with FUO, which in the past was assessed by planar ⁶⁷Ga-citrate or ¹¹¹In-labelled white blood cell (¹¹¹In-WBC) scintigraphy.⁸⁴ In a prospective study that compared FDG-PET and ⁶⁷Ga-citrate imaging in 58 patients, FDG-PET successfully detected the source FUO in 35% of cases, while ⁶⁷Ga-citrate was helpful in 25% of the subjects examined.⁸⁷ Since all ⁶⁷Ga-citrate positive cases were also visualized by FDG-PET, the authors concluded that the latter could replace other techniques for this purpose. A prospective study of 23 patients found that FDG-PET had a sensitivity and specificity of 86% and 78% in determining the etiology of FUO, compared to 20% and 100% by ¹¹¹In-WBC scintigraphy.⁸⁸ A meta-analysis comparing the utility of FDG-PET/CT, FDG-PET, ⁶⁷Ga-citrate, and ¹¹¹In-WBC scintigraphy in evaluating FUO found that FDG-PET/CT had the highest sensitivity of 86% and the highest diagnostic yield of 58%.⁸⁹

Among applications of FDG-PET/CT for the assessment of bone and soft-tissue infections, that of diabetic foot has been studied the most over the past two decades.^90–96 The possibility of diabetic neuropathy resulting in neuropathic osteoarthropathy makes diagnosis of osteomyelitis in this setting difficult by traditional clinical and radiographic techniques.^93,97 In addition to its utility in assessing FUO with suspected osteomyelitis, FDG-PET/CT has also been shown to be able to differentiate osteomyelitis from neuropathic osteoarthropathy and soft-tissue infections.^90–94,96 Besides FDG, other tracers such as NaF, ¹¹C-methionine, ⁶⁸Ga-citrate, ¹¹C-PK11195, and ¹²⁴I-FIAU have been used with PET imaging to detect and characterize osteomyelitis.^97,98 Alternatively, PET imaging with FDG-labeled autologous leukocytes has been reported to have a sensitivity of 83.3% and specificity of 100% in the diagnosis of osteomyelitis of the foot in patients with diabetes.^99,100 However, the role of the latter approach is very questionable and should not be considered for future clinical and research activities.

With increased validation for PET in the diagnosis of musculoskeletal infections and other differential causes of FUO, it becomes imperative that TB PET/CT be utilized in this domain. The nature of FUO often indicates that the underlying cause cannot be detected and localized by conventional means, and therefore only imaging of the entire body would be ideally suited to examine patients with this diagnosis. Although FDG will play a critical role in the setting of musculoskeletal infections, other PET tracers have shown some promise⁴⁹ and may also benefit from the unique capabilities of TB PET/CT.

Osteopenia and Osteoporosis

Osteoporosis, a disease characterized by increased osseous fragility and risk of fracture, has reached epidemic proportions. It affects 10 million Americans, resulting in two million fragility fractures per year.¹⁰¹ Osteoporosis is known to be a systemic disease, and age-related bone fractures could happen in any part of the skeleton. Vertebral fractures are the most common fragility fractures. Of all the osteoporotic fractures, hip fractures have the most devastating^102,103 consequences. The mortality rate in the year following a hip fracture is as high as 30%. Less than half of patients who had a hip fracture regain their previous level of function.^103,104 Other skeletal sites for fragility fractures include the wrist, pelvis, humerus, ankle, and foot.

The current standard-of-care test for the assessment of metabolic bone diseases, including osteoporosis, involves the use of dual energy X-ray absorptiometry (DXA).¹⁰⁵ DXA provides a measure of areal bone mineral density (BMD) at the hip or lumbar spine. Severity of the disease is typically quantified by the BMD T-score, which represents the standard deviations above or below the mean of a young healthy population. According to the World Health Organization (WHO) a T-score between −1.0 and −2.5 is defined as osteopenia and less than −2.5, osteoporosis.^106,107 DXA-derived BMD, a surrogate for bone quality, has many limitations, including the two-dimensional nature and poor image quality. Also, the evaluation of BMD by DXA can be confounded by osteophytosis, resulting in erroneously high quantitative values.¹⁰⁸ As a consequence, over 50% of women who sustain a hip fracture have BMD T-scores > −2.5, i.e., above the threshold for osteoporosis diagnosis and treatment.¹⁰⁹ Three-dimensional imaging modalities such as MRI^110,111 and CT^112–114 have been shown to have better performance in assessing metabolic bone diseases, particularly the bone strength, compared to DXA. However, the above mentioned imaging modalities provide only structural information about bone. Since bone is a slow-changing organ at the macro and micro level, structural changes induced by age or disease or in response to therapy cannot be detected using conventional bone imaging modalities until several months to years have passed.

Limitations in current bone imaging modalities have resulted in the exploration of novel methodologies for better assessment of bone quality. In recent years, NaF-PET/CT has emerged as a modality that can quantify changes in bone health and integrity, including in benign conditions.^115–117 Affinity of NaF to bone involves the diffusion of the radiotracer through capillaries into the extracellular fluid of the bone and the exchange of the fluoride ion with a hydroxyl group in the hydroxyapatite mineral on bone surfaces during remodeling.¹¹⁸ As a result, NaF uptake can be used as a measure of osteoblast activity and hence, bone turnover.

Quantification of both bone plasma clearance and bone uptake of NaF has been employed in the assessment of bone turnover. Both the Hawkins method and the Patlak Plot method are available to determine bone plasma clearance and require dynamic imaging. Alternatively, uptake is easily determined by a static scan and is often expressed as a standardized uptake value (SUV), which represents measured activity normalized to body weight and administered dose of tracer.¹¹⁹ NaF uptake at the femoral neck was found to decrease with age and correlated with CT-derived BMD in a study that included 68 female subjects and 71 male subjects.¹²⁰ A study of 72 postmenopausal women classified as normal, osteopenic, or osteoporotic determined by DXA demonstrated decreased lumbar spine plasma clearance of NaF in osteoporotic subjects compared to both normal and osteopenic subjects.¹²¹ Another study using NaF uptake showed similar findings of decreased bone turnover at the lumbar spine in patients diagnosed with osteoporosis compared to patients with a T score above −2.5.¹²² Besides osteoporosis, NaF-PET/CT has also shown utility in assessing metabolic bone diseases,including Paget disease of bone.^123,124

Evaluating the effects of treatment for osteoporosis is a major potential application of NaF-PET/CT. Antiresorptive therapy with bisphosphonates such as alendronate and risedronate is known to rapidly decrease bone resorption, followed by a decrease in bone formation. Accordingly, treatment of 18 women with risedronate resulted in an 18% decrease in lumbar spine plasma clearance of NaF after six months.¹²⁵ Similarly, in a study of 24 postmenopausal women with glucocorticoid-induced osteoporosis treated with alendronate, NaF uptake decreased by 14% in the lumbar spine and by 24% in the femoral neck after 12 months.¹²² Frost et al. assessed changes after discontinuation of alendronate and risedronate on NaF uptake in the spine, hip, and femur of 20 postmenopausal women.¹²⁶ The authors found that although bone turnover in the spine remained suppressed, uptake in the hip and femur increased after 12 months in patients who had taken alendronate. Treatment with teriparatide results in an increase in both bone resorption and bone formation. In 18 postmenopausal women treated with teriparatide for six months, plasma clearance of NaF at the spine as well as uptake at the femoral shaft and pelvis were observed to increase significantly.¹²⁷

In addition to the spine, pelvis, and femur, NaF-PET/CT has also been used to assess bone formation at the parietal bone, humerus, sternum, and tibia.^128,129 The calcaneus is a bone that presents with increased fracture risk due to osteoporosis. Although there have been several studies using MRI to assess calcaneal involvement in osteoporosis,^130,131 no studies have used NaF-PET/CT yet for this purpose. TB PET/CT would ensure that all areas of concern such as the calcaneus can be assessed for low bone turnover and therefore risk for fragility fracture. Using CT segmentation, metabolic activity in the whole skeleton can be quantified on PET (Figure 5). Zirakchian Zadeh et al. used this methodology to measure NaF uptake in the whole skeleton of multiple myeloma patients to assess changes in bone turnover after therapy.¹³² A similar methodology could be applied to TB PET/CT images to determine systemic bone turnover in the entire skeleton in addition to the assessment of individual bones.

Figure 5.

Whole-body FDG-PET/CT images of a 60-year-old man with multiple myeloma. The cortical bone and bone marrow were segmented using a growing region algorithm based on Hounsfield units, followed by smoothing and closing algorithms (OsiriX software; Pixmeo SARL; Bernex, Switzerland). The global SUV_mean, which represents the whole bone marrow activity, before initiating treatment (A) was 2.02 and decreased to 1.10 after finishing the course of treatment (B).

(From Raynor WY, Al-Zaghal A, Zadeh MZ, Seraj SM, Alavi A. Metastatic Seeding Attacks Bone Marrow, Not Bone: Rectifying Ongoing Misconceptions. PET Clin. 2019;14(1): 135-144; with permission)

Due to scan time and radiation dose restrictions, conventional bone imaging modalities typically focus on one anatomic site for the assessment of metabolic bone diseases. Although osteoporosis and osteopenia are known to be systemic in nature bone quality at one skeletal site does not reflect the bone health at another site.¹³³ It is also known that bone metabolism is skeletal site dependent.¹³⁴ It is therefore useful to have the capability to assess the bone quality at all skeletal sites that are susceptible to osteoporotic fracture. TB PET/CT provides a unique opportunity to assess bone metabolism in the entire skeleton in one scan, and in the same phase of radiotracer distribution and uptake. In short, faster scan time, lower dose, and entire body coverage achievable by TB PET/CT could enable earlier detection of metabolic bone diseases before structural changes in bone could be detected by other imaging modalities.

Skeletal Muscle

The human body consists of over 500 skeletal muscles primarily responsible for contraction and relaxation and supporting other tissues of the skeletal system.¹³⁵ Skeletal muscles are distributed across the entire body, and play a crucial role in physical performance, glucose homeostasis and other metabolic functions.^136,137 Therefore, skeletal muscle loss or dysfunction can have severe health consequences, ranging from functional disability and institutionalization¹³⁸ to insulin resistance, metabolic syndrome, and obesity.¹³⁹ Sarcopenia, broadly defined as clinically significant loss of muscle mass and function, is now recognized as a hallmark of ageing.¹⁴⁰ Besides ageing, sarcopenia can also occur in other conditions such as autoimmune arthritis, heart failure, and cancer (commonly referred to as cachexia).¹⁴¹ Sarcopenia presence increases both risk for hospitalization and cost of care during hospitalization,¹⁴² with a two-fold or more increase in overall cost compared to those without the condition.¹⁴³ Sarcopenia is recognized as a muscle disease with an ICD-10-CM code that can be used to bill for care in some countries.¹⁴⁴

To date, anatomical imaging methods, such as DXA, CT, MRI, and ultrasound have been used to assess surrogate measures of muscle mass and quality of relevance to sarcopenia.^140,145 While muscle mass can be reliability estimated using these methods (via detailed image segmentation), there is no consensus on the metrics of muscle quality that best represent prognostic value.¹⁴⁶ For example, the degree of fat infiltration in muscle changes the latter’s Hounsfield unit (HU) value in CT images and may provide an assessment of muscle quality; however, there is no standardization regarding CT thresholds to establish sarcopenia diagnosis across the different conditions.^146,147 Furthermore, these anatomical imaging modalities are unable to assess the molecular activity of skeletal muscle. On the other hand, PET has been shown to be sensitive to muscle blood flow, protein synthesis, and glucose metabolism, and has been utilized in the assessment of skeletal muscle with applications ranging from diabetes¹⁴⁸ and cancer-related dysfunction,^149,150 to exercise physiology.^151,152 Studies suggest that PET may provide information regarding skeletal muscle quality that is unattainable from structural imaging modalities.^149,153 Commonly used PET radiotracers in the published literature to evaluate skeletal muscle are ¹⁵O-water,^{154–156 11}C-methyl-methionine¹⁵⁷ and FDG.^158–160 A number of other radiotracers targeting key pathological processes underlying sarcopenia are under development or are being explored in early-stage human studies.^161–163

Currently, PET/CT has a limited role in the clinical assessment of sarcopenia. However, TB PET/CT could overcome shortcomings of the existing methods. First, sarcopenia is not limited to just a few muscles,^164,165 meaning that no single muscle or muscle group is a robust representative of the systemic burden of sarcopenia.¹⁶⁶ TB PET/CT enables the assessment of the entire skeletal muscle in the body in the same phases of radiotracer distribution and uptake, and can therefore contribute towards a more comprehensive and global analysis of sarcopenia (Figure 6). Second, TB PET/CT could play a critical role in the longitudinal monitoring of sarcopenia, as the cumulative doses will be significantly lower and image acquisition is relatively fast. Third, the combination of higher spatial resolution and geometric sensitivity of TB PET/CT compared to current scanners will enable delineating activity of the different muscle groups to better understand sarcopenia pathogenesis,¹⁶⁷ and heterogeneity of radiotracer uptake,^152,168 which may provide insights for intervention. Lastly, TB PET/CT measures are quantitative and therefore provide reliable and reproducible data compared with those from physical examination and structural imaging.

Figure 6.

Maximum intensity projection (MIP) of the CT image (A) showing a methodology used to segment muscle. Two lines (horizontal parallel green lines) corresponding to 5 cm above the intercondylar notch and 5 cm below the greater trochanter were manually delineated according to pre-determined anatomical criteria. A growing region algorithm with lower and upper thresholds of 1 and 150 Hounsfield units, respectively, was used to segment the muscle (OsiriX software) (B). Applying this methodology to 71 subjects, thigh muscle volume was found to decrease with age, and uptake of FDG uptake was found to be significantly higher on the right side compared to the left.

(From Kothekar E, Yellanki D, Borja AJ, et al. 18F-FDG-PET/CT in measuring volume and global metabolic activity of thigh muscles: a novel CT-based tissue segmentation methodology. Nucl Med Commun. 2020;41(2):162-168.; with permission)

In summary TB PET/CT measures may usher in several new biomarkers of sarcopenia that will be helpful both for understanding the underlying pathogenesis of the condition and for the evaluation of and monitoring of interventions and therapy.

Neoplastic Musculoskeletal Diseases

Bone and soft tissue cancers of the musculoskeletal system are relatively uncommon but have significant impact on quality of life and are commonly associated with high morbidity and mortality in the affected population. These malignancies commonly metastasize to distant organs which may not be successfully detected and monitored by conventional imaging techniques. Whole-body PET in this setting has the advantage of imaging both the primary lesion and the metastatic sites with a single image acquisition procedure.^169–171 Hybrid imaging with PET/MRI provides certain advantages over PET/CT, as PET/MRI allows higher reading confidence compared to PET/CT.

Most malignant tumors of the musculoskeletal system demonstrate increased FDG uptake due to their high glycolytic activity compared to the surrounding normal tissue structures. FDG-PET not only allows differentiation of soft tissue and osseous lesions that cannot be fully defined by structural imaging modalities alone, but is also a key component for staging/restaging of the disease and treatment planning (Figure 7).^171–173 Erfanian et al. reported a higher accuracy for PET/MRI compared to MRI alone in delineating soft tissue malignancies.¹⁷² In a recent study by Cleary et al., who examined patients with osteosarcoma, 81% were noted to have suspicious popliteal lymph nodes on the initial MRI. However, in contrast to MRI, only one node was presumed to be metastatic based on its increased metabolic activity on PET/CT, a finding that was further confirmed in the 12 months follow-up assessment.¹⁷¹

Figure 7.

A 58-year-old man with toe pain was evaluated with MRI of the foot. A marrow-replacing lesion of the first distal phalanx with T₂ hyper- (A) and T₁ hypo-signal intensity (B) was identified, the tissue sampling of which was consistent with osteosarcoma. After surgical resection and several courses of chemotherapy (C), the patient returned for restaging with whole-body FDG-PET/CT. A hypermetabolic right ankle (Kager fat pad, D, E) and inguinal lymph nodes (D) with respective SUV_max of 4.4 and 10.4 were identified, in keeping with metastatic lymphadenopathy.

PET is commonly employed for follow-up of aggressive bone and soft tissue tumors to evaluate therapy response and is often superior to structural imaging alone for this purpose. In fact, molecular imaging is the mainstay of personalized therapeutic approaches in musculoskeletal cancer treatment. A recent study reported by Lee et al. described the impact of PET/CT in managing 73 patients with stage II osteosarcomas of the extremities who were treated with two cycles of neoadjuvant chemotherapy, surgical resection, and adjuvant chemotherapy. These patients underwent PET/CT before treatment (PET0), after 1 cycle of chemotherapy (PET1), and following the completion of neoadjuvant chemotherapy (PET2). They noted that evidence for response to treatment after the first cycle of preoperative chemotherapy was best predicted by PET.¹⁷⁴ This research demonstrated that PET was a powerful modality for early response monitoring by providing the opportunity of early modification of timing of local control. SUV_max on PET2, the delta (percentage change) of SUV_max between PET0 and PET1, and between PET0 and PET2 have been proposed by these researchers as the most accurate predictors of poor response and development of future metastatic events. Based on the data generated, patients with SUV of more than 5.9 on PET2 had a poor event-free survival compared to the others.¹⁷⁴ The features provided by PET have been further augmented by the incorporation of artificial intelligence, texture analysis, and machine learning into image interpretation.^175,176

Despite all the advantages associated with this modality, some studies have warned about false negative results with PET imaging. A retrospective study on 24 histopathologically confirmed cases of Ewing sarcoma who underwent MRI and FDG-PET/CT within a 4-week interval demonstrated that some of the osseous metastases detected by MRI may not show increased metabolic activity on FDG-PET/CT, rendering them as false-negative results.¹⁷⁷ Therefore, it has been advised that caution should be exercised in interpreting FDG-PET/CT for the detection of skeletal metastases in Ewing sarcoma. Likely, the potential source of such false negative results is due to the presence of low contrast resolution between bone metastases and the background hematopoietic marrow, hematopoietic reconversion due to recent chemotherapy, and/or due to the small size of the lesions that are not detectable by PET. In contrast, MRI may not be significantly affected by these factors.¹⁷⁷ We believe these findings emphasize the importance of integrated PET/MRI in these settings by combining the advantages of molecular imaging with those of MRI by providing structural details.^5,178,179 Also, it has been widely accepted that non-contrast chest CT is the modality of choice for the evaluation of pulmonary metastases of osteosarcoma and Ewing sarcoma due to suboptimal sensitivity of PET/CT in detecting lung nodules that are smaller than 2 cm and are located in the lower lobes (due to technical limitations of respiratory motion gating and misregistration of PET and CT scans).^180–182 However, TB PET/CT imaging may overcome some of these shortcomings by acquiring images over a shorter period of time and at later time points (at 3-4 hours).

Discussion

As PET-based techniques become increasingly validated, their application to MSDs, such as arthritis, infection, osteoporosis, sarcopenia, and soft tissue and osseous neoplastic pathologies, will continue to grow. Many of these disorders are difficult to diagnose, stage, and evaluate over time with conventional imaging modalities. FDG and NaF among other tracers have the potential to reveal molecular processes before evidence of disease is present clinically or structurally on conventional imaging techniques. Although reported studies in the literature are based on examining individual joints, bones, and muscles, TB PET/CT will allow for a comprehensive and systematic evaluation of the MSDs that cannot easily be localized to one specific region. The introduction of TB PET/CT in this domain may allow for improved detection of many abnormalities throughout the body, accurate monitoring, and a better understanding of the effects of the current and future therapeutic intervention.

However, before TB PET/CT can be widely adopted, certain aspects of this modality require further consideration and optimization. Regarding chronic conditions as well as pediatric applications, ionizing radiation dose may be a concern, and a cost-to-benefit analysis would be warranted to determine optimal scan frequency. Furthermore, depending on the kinetics of the radiotracer, the preferred scanning start time and duration may be disease-specific, causing difficulty in visualizing unrelated findings (such as evaluating cardiovascular disease in RA). Lastly, reconstruction parameters will also depend on the structure of interest. For example, imaging of small joints in RA would warrant parameters that produce high resolution, but high variance, while imaging of large muscles would warrant parameters that produce low variance, but lower resolution. Future research regarding the tradeoffs involved in each of these considerations would support the efficient use of TB PET/CT and facilitate wider applications of this new but very promising technology.

Key Points (3-5 bulleted sentences indicating the main takeaways/defining elements of the article).

• A number of musculoskeletal disorders are systemic in nature or have systemic sequalae. Biomarkers that provide global assessment of disease activity across the entire body are urgently needed.

• Total-body PET/CT can provide anato-molecular imaging-based measures across the entire human body at reduced dose, lower scan time, and higher spatial resolution.

• Total-body Pet/CT measures can contribute towards an improved understanding of the underlying pathogenesis of musculoskeletal disorders and may provide new insights for the evaluation of and monitoring of interventions and therapy.

• Since total-body PET/CT includes images of the brain in the field of view, this approach may provide a means to detect and quantify the degree of pain that is associated with musculoskeletal disorders, potentially allowing further insight into the impact of opioids.

Synopsis.

Imaging of musculoskeletal disorders, including arthritis, infection, osteoporosis, sarcopenia, and malignancies, is often limited when using conventional modalities such as radiography, CT, and MRI. As a result of recent advances in PET instrumentation, total-body PET/CT offers a longer axial field-of-view, higher geometric sensitivity, and higher spatial resolution compared to standard PET systems. This article discusses the potential applications of total-body PET/CT imaging in the assessment of musculoskeletal disorders.