Comparing gene expression profiles of adults with isolated spinal tuberculosis to disseminated spinal tuberculosis identified by 18FDG-PET/CT at time of diagnosis, 6- and 12-months follow-up: classifying clinical stages of spinal tuberculosis and monitoring treatment response (Spinal TB X cohort study)

Background Tuberculosis (TB) is one of the top ten causes of death worldwide, with approximately 10 million cases annually. Focus has been on pulmonary TB, while extrapulmonary TB (EPTB) has received little attention. Diagnosis of EPTB remains challenging due to the invasive procedures required for sample collection. Spinal TB (STB) accounts for 10% of EPTB and often leads to lifelong debilitating disease due to devastating spinal deformation and compression of neural structures. Little is known about the extent of disease, although both isolated STB and a disseminated form of STB have been described. In our Spinal TB X cohort study, we aim to describe the clinical phenotype of STB using whole-body 18FDG-PET/CT, identify a specific gene expression profile for different stages of dissemination and compare findings to previously described gene expression signatures for latent and active pulmonary TB. Methods A single-centre, prospective cohort study will be established to describe the distributional pattern of STB detected by whole-body 18FDG-PET/CT and gene expression profile of patients with suspected STB on magnetic resonance imaging (MRI) at point of diagnosis, six months, and 12 months. Blood biobanking will be performed at these time points. Specimens for microbiology will be obtained from sputum/urine, from easily accessible sites of disease (e.g., lymph nodes, abscess) identified in the first 18FDG-PET/CT, from CT-guided biopsy and/or surgery. Clinical parameters and functional scores will be collected at every physical visit. Data will be entered into RedCap® database; data cleaning, validation and analysis will be performed by the study team. The University of Cape Town Ethics Committee approved the protocol (243/2022). Discussion The Spinal TB X cohort study is the first prospective cohort study using whole-body 18FDG-PET/CT scans in patients with microbiologically confirmed spinal tuberculosis. Dual imaging techniques of the spine using 18FDG-PET/CT and magnetic resonance imaging as well as tissue diagnosis (microbiology and histopathology) will allow us to develop a virtual biopsy model. If successful, a distinct gene-expression profile will aid in blood-based diagnosis (point of care testing) as well as treatment monitoring and would lead to earlier diagnosis of this devastating disease. Trial registration: The study has been registered on ClinicalTrials.gov (NCT05610098). Supplementary Information The online version contains supplementary material available at 10.1186/s13018-024-04840-7.


Discussion
The Spinal TB X cohort study is the first prospective cohort study using whole-body 18FDG-PET/CT scans in patients with microbiologically confirmed spinal tuberculosis.Dual imaging techniques of the spine using 18 FDG-PET/CT and magnetic resonance imaging as well as tissue diagnosis (microbiology and histopathology) will allow us to develop a virtual biopsy model.If successful, a distinct gene-expression profile will aid in blood-based diagnosis (point of care testing) as well as treatment monitoring and would lead to earlier diagnosis of this devastating disease.

Epidemiology
In 2022, an estimated 10.6 million people fell ill with tuberculosis (TB) and 6.4 million patients received TB treatment [1].TB and HIV are the two leading infectious diseases causes of death worldwide, with TB ranking number one cause of death in people living with HIV (PLWH) and responsible for the burden of multimorbidity in low-and-middle-income countries [2][3][4].South Africa has the most devastating TB/HIV epidemic with an HIV prevalence of 19% in adult TB cases and a TB incidence of 615/100,000 in some regions [5][6][7][8].In 2019, 7.1% of the previously treated TB cases and 3.4% of the new TB cases in South Africa were resistant to rifampicin and/or isoniazid [9].TB vaccine development has recently been successful, but not in PLWH who are more likely to develop extrapulmonary TB (EPTB) [10][11][12].Musculoskeletal TB accounts for approximately 20% of EPTB cases, with spinal involvement in up to 50 percent of these cases (Fig. 1) [13,14].It is believed that spinal tuberculosis (STB) arises from the hematogenous spread of Mycobacterium tuberculosis (Mtb) from a primary pulmonary infection into the well perfused cancellous bone of the vertebral bodies.[15] STB is a spondylitis or spondylodiscitis caused by Mtb, historically called Pott's disease or Pott`s spine.It commonly manifests as localised backpain, pain, spinal deformity with/without instability, neurological deficit, and constitutional symptoms.The time from initial symptoms to diagnosis can take several years [16][17][18].Diagnosis relies on clinical presentation, spinal imaging, and microbiological testing from lesion biopsy [19].Therapy includes TB treatment, immobilization, and spinal surgery in selected patients [20].

Clinical and radiological phenotypes of spinal TB
Little is known about the extent of disease in STB, and isolated STB, which is confined to the vertebra and surrounding anatomical structures, as well as disseminated forms of STB have been described (Fig. 2, 3) [21,22].The latter presents with spinal lesion(s) plus additional TB lesion(s) in other parts of the body such as the lungs, pleura, lymph nodes, meninges, urogenital, gastrointestinal tract, and other musculoskeletal involvement such as peripheral joints [23].The diagnosis of disseminated STB requires whole body imaging and/or microbiological evidence of additional sites of infection, e.g., a positive GeneXpert from a lymph node aspirate or a positive urine culture for Mtb [24].With this, the question arises, whether isolated spinal TB without any other active focus can be a stand-alone entity and whether the disseminated form of STB is a dual Mtb infection with potentially different strain types at the different sites of disease [25].

Imaging in tuberculosis
Magnetic resonance imaging (MRI) is the imaging gold standard for STB and has a reported sensitivity of 94% and specificity of 93% in detecting spondylodiscitis [26].MRI can accurately detect spinal cord compression, intrinsic changes of the cord, bony changes of the vertebral body as well as the extent of disc destruction, and psoas abscess formations [27].If possible, MRI of the whole spine should be performed to detect noncontinuous lesions, occurring in up to 20% of patients [28]  been shown to be able to detect sites of infection and monitor treatment response in tuberculosis [29]. 18FDG accumulates in tissue with increased glucose metabolism (e.g., infections, inflammatory conditions, malignant conditions, ischemia).It has been found that 18 FDG uptake values are higher in patients with STB compared to patients with pyogenic spondylitis [30].A small study has shown that 18 FDG-PET/CT is potentially superior to MRI in accuracy to diagnosing STB [31].Semi-automatic technique to quantify complex tuberculous lesions on 18 FDG-PET/CT and allows for standardised assessment [32].

Transcriptomic profiling in tuberculosis
Host transcriptomics have identified RNA signatures for tuberculosis diagnosis and monitoring treatment response as well as understanding the immunological mechanisms of pulmonary TB [33,34].Several RNA signatures have been identified for risk of disease stratification, screening for TB, tracking treatment response and prediction of treatment failure [35][36][37][38].Lately, an RNA signature which predicts recent exposure to Mtb in humans has been identified [39].The description of these RNA signatures is opening the possibility for developing large-scale point of care screening tests as well as targeted (personalized) intervention to prevent the disease and, in patients with active tuberculosis, create customized treatment plans [36].However, the diagnosis of EPTB and particularly STB remains challenging simply because sample collection requires invasive procedures in the absence of a blood-based diagnostic test.Host transcriptomics may be the key for a blood-based test for STB.

Treatment of spinal TB
Treatment combination and duration of treatment of STB has not been systemically studied in randomised controlled trials for various reasons.According to WHO, adults with extrapulmonary TB can safely be treated with a 6-month regimen except for patients with TB of the central nervous system and musculoskeletal TB, including patients with STB.These subgroups require 9-12 months of treatment duration.
Treatment for STB consists primarily of chemotherapy (antituberculosis treatment; ATT) and in case of a "spine at risk", surgery is required additionally [40,41].Drug-sensitive TB can be treated using the standard combined 4-drug regimen medication according to body weight.The WHO suggests an intensive phase using with rifampicin, isoniazid, pyrazinamide, and ethambutol for two months, followed by continuation phase using rifampicin and isoniazid for seven to ten months [42] Guidelines for the treatment of multi-drug resistant (MDR) STB are lacking.According to WHO, patients with severe forms of extrapulmonary MDR TB including TB meningitis, brain abscesses, osteomyelitis and arthritis (which includes STB) should not be treated with the new 9-month all-oral regimen for MDR-TB and longer regimens apply for patients with severe forms of extrapulmonary diseases such as STB [43].However, treatment durations of STB as well as dosing of TB drugs differ significantly between different countries and an infectious diseases expert with experience in extrapulmonary TB disease should always be consulted [44].

Aims and objectives
The aims of this project are to (1) describe the clinical phenotype of STB (isolated versus disseminated STB), (2) explore different imaging modalities using MRI of the spine and whole-body PET/CT, (3) identify RNA signatures for the two different clinical phenotypes of STB: the isolated and disseminated form of STB, and (4) to analyse the genomes of Mtb isolates extracted from different specimens from different sites of the body, hence identifying dual infection.Objectives are listed in Box 1.

Study design, setting, and eligibility
This is a prospective cohort study of HIV-infected and -uninfected patients with STB who live in the Cape Town metropole, South Africa.Patients with suspected STB will be referred from Groote Schuur Hospital and its affiliated teaching hospitals (Mitchell`s Plain Hospital,

Primary objective
To describe the clinical phenotype of spinal TB using whole body PET/CT and to identify mRNA gene expression profiles of isolated spinal TB versus disseminated spinal TB stratified by HIV status Secondary objectives 1.To identify the distributive patterns of suspected spinal TB using two imaging modalities: MRI and PET/CT 2. To analyse the genomes of Mtb isolates from different sites of the body (if available) and to identify differences in their genome regarding SNPs and drug susceptibility 3. To analyse imaging findings using PET/CT at treatment initiation, 6 months, and 12 months to better understand PET/CT as an imaging quality for the judgment of treatment outcome 4. To compare imaging findings on PET/CT and MRI at baseline to evaluate the role of PET/CT in spinal TB diagnostics (specific semi-quantitative PET/CT values as virtual biopsy) New Somerset Hospital, Victoria Hospital Wynberg) to the study team at the University of Cape Town.Eligibility criteria are listed in Box 2. We aim to recruit 100 participants over a timeframe of three years.

Patient and public involvement
Patients or the public are not involved in the design, or conduct, or reporting, or dissemination plans of this prospective cohort study.

Case definition
Only cases with clinical signs and symptoms and suspicion of STB on MRI will be included in the Spinal TB X cohort study (Fig. 4).Definite STB is defined as microbiological evidence of Mtb infection in tissue of spinal lesions or psoas abscess drainage.

Study procedures
Detailed study procedures and timelines are displayed in Table 1 and Fig. 5.At the pre-screening visit, patients will be assessed regarding their eligibility to join the study.Routine clinical work-up includes blood tests, MRI and work-up for tuberculosis.Patients with suspected spinal tuberculosis on MRI will be approached to join the Spinal TB X cohort study.Informed consent will be obtained by study nurses in the native language of the patient.
After inclusion in the study, the screening visit will be performed.At baseline, clinical examination as well as blood collection [creatinine, Alanine transaminase (ALT), full blood count (FBC) including differential count (DIFF), C-reactive protein (CRP) and Haemoglobin A1C (HbA1c)] will be performed.CD4-count, and HIV viral load will be tested in PLWH.Every patient with no history of HIV infection will undergo HIV testing according to national guidelines.Sputum and urine will be collected for GeneXpert Ultra test and Mtb culture including drug sensitivity testing, and screening for diabetes and pregnancy will be completed.Whole-body PET/CTs will be performed within ten days after the screening visit.At the PET/CT visit, serum, heparin blood (for Peripheral Blood Mononuclear Cells) and whole blood (in PAXgene tubes) will be collected for biobanking.After the PET/CT, patients will undergo spinal biopsies, or surgery where applicable and specimen will be collected for GeneXpert Ultra test and Mtb culture including drug sensitivity testing.Within ten days of the PET/CT visit, patients will be reviewed and examined by the study team and PET/CT findings will be discussed.Where applicable, patients will be referred to medical specialties for further biopsies of sites of infection detected by PET/CT.Patients with microbiologically confirmed "definite" STB on the collected specimen will be started on TB treatment according to local guidelines.Patients with other diagnoses confirmed by culture or histology will be excluded from the study.In the following one to five months, every patient will be contacted by the study team on a regular basis and undergo a standardized questionnaire regarding their clinical course of disease and tuberculosis and HIV (if applicable) drug adherence monitoring.Six months after the first PET/CT, patients will undergo the second PET/CT.At the second PET/CT visit, serum, heparin blood (for Peripheral Blood Mononuclear Cells) and whole blood (in PAXgene tubes) will be collected for biobanking.Additionally, FBC including DIFF and CRP will be collected to determine inflammatory trends.Furthermore, a random-glucose test and in female participants, a beta-HCG-urine test, will be performed.At months seven to eleven, telephonic follow-up will be continued and at month 12, PET/CT 3 will be performed.At the third PET/CT visit, serum, heparin blood (for Peripheral Blood Mononuclear Cells) and whole blood (in PAXgene tubes) will be collected for biobanking.Additionally, FBC including DIFF and CRP will be collected to determine inflammatory trends.Furthermore, a random-glucose test and in female participants, a beta-HCG-urine test, will be performed.

Data collection, data cleaning, and statistical analysis
All study source data will be collected on paper case report forms and entered and stored in RedCap ® on a dedicated secure central database at the University of Cape Town.Data will be reviewed by at least two investigators (JS, FT, SM, KW) for completeness and data cleaning, validation and analysis performed by the academic study team.All data will be transferred to R, Version 4.2.2 for all analysis.Normally distributed continuous data will be presented as mean ± SD and non-Gaussian distributed variables as median + IQR.Categorical data will be presented as percentages with 95% CI where appropriate.For patient group comparisons, we will use χ 2 analysis with calculation of ORs and 95% CI (where appropriate) for discrete variables and student t-test and analysis of variance for normally distributed continuous variables.Multiple logistic regression analyses (entry model) will be performed on age, sex, and baseline characteristics where applicable.
Significance will be accepted at the two-sided level of 0.05.Whole-genome and RNA-sequencing will be performed through third parties.Measurement of semiquantitative PET/CT values will be performed using MIM ® Version 7.2.8 (MIM Software Inc., Cleveland,

Fig. 2 Fig. 3
Fig. 2 Definition of isolated spinal tuberculosis (TB) and disseminated spinal TB in the Spinal TB X cohort

Fig. 4
Fig. 4 Case definition for "definite" spinal tuberculosis (TB) in the Spinal TB X cohort.MRI magnetic resonance imaging, Mtb Mycobacterium tuberculosis, AFB acid-fast bacilli (Ziehl-Neelsen or auramine-rhodamine stain), CT computed tomography Any person for whom the physician feels this study is not appropriate (e.g., patients already on TB treatment)