Can FDG PET/CT optimize therapy in patients with drug-sensitive pulmonary tuberculosis in patients who have completed a standard course of anti-tuberculous therapy?

Introduction: Pulmonary residual metabolic activity (RMA) is prevalent on FDG PET/CT imaging of patients who completed a standard course of anti-tuberculous therapy (ATT) for pulmonary tuberculosis (PTB). RMA has an association with tuberculosis relapse. We aimed to evaluate the impact of RMA on EOT FDG PET/CT in patients with PTB after completing a standard course of ATT.

Methods: Approval for this study was granted by the Ethics Committee of the University Hospital of Pretoria.We prospectively recruited patients who were declared cured based on clinical and microbiological assessment after completing a standard course of ATT . All patients subsequently underwent [¹⁸F]F-FDG PET/CT within two weeks of completing ATT. We determined the proportion of patients who had RMA in the lungs as well as CT features suggestive of active PTB including lung consolidation, pulmonary nodules, lung changes in tree-in-bud pattern, hilar/mediastinal lymphadenopathy, and pleural effusion. All patients were followed up for 6 months to evaluate for relapse. Confirmatory bacteriologic testing was performed in patients with suspected relapse who demonstrated symptoms. In those who relapsed, a repeat FDG PET/CT imaging was subsequently performed.

Results: We studied 75 patients including 50 HIV-infected individuals with a mean age of 36.09 ± 10.49. Median CD4 count among HIV-infected patients was 255 cells/mm³ (IQR: 147– 448). HIV viral load was 12497 copies (IQR:158-38841). HIV-infected patients had lower hemoglobin levels (13.07±1.78 g/dL versus 14.24±2.07, p=0.021) and higher C-reactive protein levels (5.70 versus 1.20, p=0.001) compared with HIV-uninfected patients. All other baseline clinical and demographic characteristics were not significantly different between the groups. Forty-one patients had RMA and its incidence was not significantly different between HIV-infected and uninfected patients (p=0.101).Thirty-four patients demonstrated complete metabolic response (CMR) to ATT. In the RMA group, 3 patients relapsed on follow-up, whereas no tuberculosis relapse was demonstrated in those with CMR. Among the four CT features of active PTB, only metabolically active nodes were more prevalent among HIV-infected compared with uninfected patients (50.0% versus 5.9%, p=0.003) while the incidence of the other three CT features of active PTB was not significantly different between HIV-uninfected patients with RMA versus HIV-uninfected patients (P>0.05 in all cases).

Conclusions: FDG PET/CT imaging demonstrates a high prevalence of RMA among patients treated with a standard course of ATT for PTB, with a similar incidence between HIV-infected and uninfected patients. The risk of relapse within 6 months after completing ATT in patients with CMR is negligible. This study further iterates the potential of FDG PET-CT imaging in determining the optimal treatment duration in the management of patients with drug-sensitive PTB. However, tuberculosis relapse may be seen in 9% of patients within 6 months of completing ATT. The incidence of CT lung changes suggestive of active PTB was similar between the groups, indicating that the presence of HIV coinfection may not influence the interpretation of EOT FDG-PET/CT obtained for PTB treatment response assessment.