Leprosy reactions: The predictive value of Mycobacterium leprae-specific serology evaluated in a Brazilian cohort of leprosy patients (U-MDT/CT-BR)

Abstract

Background: Leprosy reactions, reversal reactions/RR and erythema nodosum leprosum/ENL, can cause irreversible nerve damage, handicaps and deformities. The study of Mycobacterium leprae-specific serologic responses at diagnosis in the cohort of patients enrolled at the Clinical Trial for Uniform Multidrug Therapy Regimen for Leprosy Patients in Brazil/U-MDT/CT-BR is suitable to evaluate its prognostic value for the development of reactions.

Methodology: IgM and IgG antibody responses to PGL-I, LID-1, ND-O-LID were evaluated by ELISA in 452 reaction-free leprosy patients at diagnosis, enrolled and monitored for the development of leprosy reactions during a total person-time of 780,930 person-days, i.e. 2139.5 person-years, with a maximum of 6.66 years follow-up time.

Principal findings: Among these patients, 36% (160/452) developed reactions during follow-up: 26% (119/452) RR and 10% (41/452) had ENL. At baseline higher anti-PGL-I, anti-LID-1 and anti-ND-O-LID seropositivity rates were seen in patients who developed ENL and RR compared to reaction-free patients (p<0.0001). Seroreactivity in reactional and reaction-free patients was stratified by bacilloscopic index/BI categories. Among BI negative patients, higher anti-PGL-I levels were seen in RR compared to reaction-free patients (p = 0.014). In patients with 0<BI<3, (36 RR, 36 reaction-free), higher antibody levels to PGL-I (p = 0.014) and to LID-1 (p = 0.035) were seen in RR while difference in anti-ND-O-LID positivity was borderline (p = 0.052). Patients with BI≥3 that developed ENL had higher levels of anti-LID-1 antibodies (p = 0.028) compared to reaction-free patients. Anti-PGL-I serology had a limited predictive value for RR according to receiver operating curve/ROC analyses (area-under-the-curve/AUC = 0.7). Anti LID-1 serology at baseline showed the best performance to predict ENL (AUC 0.85).

Conclusions: Overall, detection of anti-PGL-I, anti-LID-1 and anti-ND-O-LID antibodies at diagnosis, showed low sensitivity and specificity for RR prediction, indicating low applicability of serological tests for RR prognosis. On the other hand, anti-LID-1 serology at diagnosis has shown prognostic value for ENL development in BI positive patients.

Trial registration: ClinicalTrials.gov NCT00669643.

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Fig 1

Baseline antibody responses to PGL-I (A), LID-1 (B) and ND-O-LID (C) in reaction-free (n = 292) and reactional leprosy patients (RR+ENL, n = 160). The box shows the interval between the first and the third quartiles, the middle line represents the median. The p value refers to differences in OD medians. The traced horizontal line is the cut-off: PGL-I OD>0.25; LID-1 OD>0.3; ND-O-LID OD>0.923. The numbers above each box represent the positivity rate and the points above each box are outlier results. OD = optical density.

Fig 2

Baseline antibody responses to PGL-I (A), LID-1 (B) and ND-O-LID (C) among reaction-free (n = 292), leprosy patients that developed RR (n = 119) and leprosy patients that developed ENL (n = 41). The box shows the interval between the first and the third quartiles, the middle line represents the median. The p value refers to differences in OD medians. The traced horizontal line is the cut-off: PGL-I OD>0.25; LID-1 OD>0.3; ND-O-LID OD>0.923. The numbers above each box represent the positivity rate and the points above each box are outliers results. OD = optical density; RR: reversal reaction; ENL: erythema nodosum leprosum.

Fig 3

Group 1- BI negative patients: Antibody responses to PGL-I (A), LID-1 (B) and ND-O-LID (C) antigens in reaction-free patients (n = 196) and patients that developed RR (n = 23). The box shows the interval between the first and the third quartiles, the middle line represents the median. The p value refers to differences in medians of OD. The traced horizontal line is the cut-off value: PGL-I OD>0.25; LID-1: OD>0.3; ND-O-LID: OD>0.923. The numbers above each box represent the positivity rate and the points above each box are outlier results. OD = optical density; RR: reversal reaction.

Fig 4

Group 2- Patients with 0p value refers to differences in OD medians. The traced horizontal line is the cut-off: PGL-I OD>0.25; LID-1: OD>0.3; ND-O-LID: OD>0.923. The numbers above each box represent the positivity rate and the points above each box are outliers results. OD = optical density; RR: reversal reaction.

Fig 5. Group 3- Patients with BI≥3 (n = 161): Baseline antibody responses to different M. leprae antigens.

Seropositivity to PGL-I (A), LID-1 (B) and ND-O-LID (C) antigens in patients that developed RR (n = 60), ENL (n = 41) and reaction-free (n = 60) patients. The box shows the interval between the first and the third quartiles, the middle line represents the median. The p value refers to differences in OD median. The traced horizontal line is the cut-off: PGL-I OD>0.25; LID-1: OD>0.3; ND-O-LID: OD>0.923. The numbers above each box represent the positivity rate and the points above each box are outliers results. OD = optical density; RR: reversal reaction; ENL: erythema nodosum leprosum.

Fig 6

Receiver Operating Curve (ROC) for serology to (A) PGL-I, (B) LID-1 and (C) ND-O-LID antigens in leprosy patients who developed RR during follow-up and reaction-free.

Fig 7

Receiver Operating Curve (ROC) for (A) PGL-I, (B) LID-1 and (C) ND-O-LID antigens in leprosy patients who developed ENL during follow-up and reaction-free.