INI I RNAI R)NAI. JOt RNAI. 01. I.P.PRUSY^

Volume 67, Number 3l'rinted in the (/..S.,1.

(ISSN 1)145-916x)

CORRESPONDENCE

This (levarnncn' is for lhe puhlication of in/mirrai cmmmunications that are ol . interest

becanse theV are infor atire and stinndating, and /or the discussion o/ c'ontrol'ersial

matters. The mandou' o/ tiús lo(e i/, is to disseminou- i/iJormatian relating to leprosa' in

particular and also other m cobaclerial descases. Dissident comutem or interpretation on

ptehlislued research is o/ conrse rulid, bui personality ntiaeks on individuais n ould secm

unnecessarv. Political commcnis, i'alid or not, also are unu'elcomc. Thev rniç'ht residi ia

interferente mth the distribution o/ the fotinAtm. and tines inter/ere with its prime pur7 xrse.

A Study on the Methods for Early Serological Diagnosisof Leprosy and Their Potential Use

To T ilk EDITOR:

As is well known, leprosy is a chronicinfectious disease caused by Mvcobac-teriam leprae. It principally affects the pe-ripheral nerves and the skin. In persons suf-fering from lcpromatous leprosy, leprosy isa generalized disease affecting several sys-tems and many organs. Recent studies haveindicated that, although multidrug therapy(MDT) has been implcmentcd, the leprosydetection rates still have not declined and insome arcas of the world have cven in-creased, as has been the case with leprosyrelapse due to drug resistance. Hence, lep-rosy is still a disease which will continueinto the next century.

in 1997, the World Health Ore*anization(WHO) Expert Committee on Leprosypointed out that continue(' leprosy researchwas needed (WHO working paper; SeventhMecting of the WHO Expert Committee onLeprosy, 26 May-3 June 1997, Geneva).According to the WHO Expert Committeeon Leprosy's research agenda, the tests tomeasure M. leprae infection is still an im-portant project. The tests for the serologicaldiagnoses of leprosy (TSDL), especiallyELISA based on phenolic glycolipid-I(PGL-I) and their artificial products of epi-topes as antigens, have been standardized.However, their practical evaluations stillhave not been determined. We have system-

atically evaluated these TSDL, and reportthe results in this conlmunicallon.

MATERIALS AND METHODSSera for detection. Sera were collected

from 644 cases with various types of lep-rosy (155 LL, 168 BL, 95 BB, 114 BT, 112TT) classifìed according to Ridley-Joplingcriteria; 160 patients with tuberculosis; 570normal controls (NC) in a nonendemic arcaof leprosy; 884 household contacts (HC),3603 random population (RP), 369 leprosypatients clinically cured with dapsone(LPD), and 8 multibacillary patients to betreated with MDT (MPM).

Antigens. Materiais used included:Whole M. smegnuttis (Ms), prepared in ourlaboratory; whole M. leprae (WML), pro-vided by Dr. Douglas; PGL-i, ND-O-BSAand ND-P-BSA, provided by Dr. Brennan;NT-O-BSA and NT-P-BSA, provided byDr. Buchanan. HRP-IgA, -IgG, -IgM and-IgAGM were provided by Dr. Douglas.FITC-IgG was provided by Dr. Abe.

The TSDL evaluated, a total of 12 tech-nologies, were as follows:

1. Fluorescent leprosy antibody absorp-tion test (FLA-ABS), according to Abe'stechnology (`).

2. Indirect enzyme-linked immunosor-bent assay (ELISA) based on PGL-I as anantigen (PGL-I-ELISA), according to Bren-nan's technology (".7).

302

67, 3^ Corre spondence^ 303

3. ELISA based on M..smegnaatis as anantigen (Ms-ELISA), according to Doug-las' technology ( 4 ).

4. ELISA based on WML as an antigen(ML-ELISA), according to Brctt's technol-ogY e).

5. ELISA based on ND-O-BSA, NT-O-BSA, ND-P-BSA and NT-P-BSA as anti-gens separately (ND-O-BSA-ELISA, NT-O-BSA-ELISA, ND-P-BSA-ELISA andNT-P-BSA-ELISA), basically according toBrennan's procedures; some parametershave becn optimized hy us ( 2 '1.

6. Serum antibody competition test(SACT), basically according to Klatscr'sand Shinha's procedures; some parametershave heen optimized hy us ( 2 ).

7. Monoclonal antibody specific hindingassay (McAb/SBA), cstahlished by Wu, Q.,et al. ( 6 ).

8. Latex agglutination test (LAT), estab-lished by Wu, Q., et al. (').

9. Gelatin particle agglutination test(GPAT) (Fujerebio, Inc., Tokyo, Japan),provided hy WHO.

Statistical analysis. The cut-off value,i.e., normal value, was determinai by thepercentile method; evaluations for variousmethods were conducted with the /N ratio,Student's t test, correlation analysis, indi-vidual agreement, sensitivity, specificity,Youden's índex, likelihood ratio (LR), etc.

RESULTSResults of 12 TSDL evaluations. A to-

tal of 12 methods were established accord-ing to the original author's proceduresand/or our procedures (mimes and proce-dures in Materiais and Methods). Resultsfor the evaluations were as follows:

From the data obtained based on param-eters of scnsitivity, specificity, practicabilityand correlation among them on a largescale, the results indicated that: a) For de-tecting the antibody levei ND-O-BSA-ELISA is the best, PGL-I-ELISA is. second.Ms-ELISA may be used as a primary toolfor screening infection with M. leprae in or-der to save specific artificial products asantigen. b) LAT, GPAT, and ND-O-BSAELISA were comparable in cfetecting re-sults; LAT and GPAT are simple, morerapid and more suitable for use in the field.c) artificial products were better than nat-

ural antigens, especially easier to coza andto standardize. d) McAh-SBA were moresuitable for detecting paucibacillary (PB)cases hecause a highcr detection rate wasobtained in PB patients.

Blood from the earlobes may be substi-tuta' for venous blood, and the blood maybe absorhed onto a fìlter paper strip anddried at room temperature (filter paper blot-ling dried blood spot); the dried spot can bereconstituted hy adding a suitable amountof water for the detection tests. Its advan-tages are: the equipment for collectingblood is simpler, the blood is easier to col-Icct, transport, store and keep the antibodyreactivity; this method is cheaper, making alargo-scale study possible.

After systematically evaluating the twokinds of inexpensive blocking agents (skimmilk and egg alhumin), we discovered theiroptimizing conditions for TSDL. They notonly increased lhe specificity and sensitiv-ity of the TSDL but they also made themmore simple and more economical.

Studies on practicability. Our studieson practicability included: 1) Sero-im-munological epidemiology. In ND-O-BSA-ELISA, the increase or decrease of the opti-cal density (OD) value has a positive corre-lation with the bacterial index (BI) and theorder of positive rates was: a) in varioustypes of lcprosy: LL>BL>BB>BT>TT; b)in household contacts (HC), random popu-lation (RP), normal controls of endemic ar-cas (ENC), and normal controls of nonen-demic arcas (NNC): HC>RP>ENC>NNC.2) In populations with subclinical M. lepraeinfection: the highest risk group was be-tween the ages of 15 and 25 with an in-crease or a persistence of high OD valuesprior to the onset of disease. 3) OD valuesgradually decrcased over time followingtreatment and these declines paralleled de-clines in the BI. 4) In clinically cured lep-rosy cases after dapsone therapy, an in-crease or a persistence of high OD values inthe ND-O-BSA-ELISA preceded the onsetof lcprosy relapse.

DISCUSSIONBasic evaluation fiar 'TSDL. For sensi-

tivity, specificity, speed, simplicity andeconomy the ND-O-BSA-ELISA and theGPAT are recommended. The former was

304^ hUernational Journal o(Leprosy^ 1999

more sensitive than the !atter. The latter ismore simple and more rapid than theformer, thus more suitable for use in thefield. The other assays can be used undcrcertain conditions. A1though nove of theTDSL could be used independently to con-fim] diagnosis or as typing tools for leprosy,each of them except the Ms-ELISA canprovicle support for the diagnosis of lep-rosy, especially in its early stages and forexcluding suspected cases, and for infectionwith other mycohacteria, as well as for fol-low up after completion of treatment, earlydetection of relapse and even for distin-guishing relapse from reversal reaction(WHO working paper; Seventh Meeting ofthe WHO Expert Committee on Leprosy,26 May-3 June 1997, Geneva). Addition-ally, TSDL can be used for sero-epidemio-logical studies to detect subclinical infec-tion with Al. leprae and for populations athigh risk, for monitoring changes in the in-tensity of M. leprae exposure, and to deter-mine epidemiological trcnds, extent, andtransmission of infection with Al. leprae.They will increase our knowledge of theepidemiology of M. leprae infection. TSDLdo not detect ali clinically apparent infec-tions because the majority of paucibacillarypatients do not demonstrate humoral re-sponse. This is an obstacle of TSDL. In or-der to solve this problem, the develòpmentof a new, more sensitive and more specificassay is very necessary.

Limitation of TSDL. TSDL of leprosyare clearly useful, although in a limitedway. The reasons that the results differ fromstucly to study may bc that: a) ali antigens'sensitivity and specificity are still not satis-factory; b) TSDL still need to be optimizedand standardized. At present, different lab-oratories in the world conduct TSDL withdifferent procedures and agents, especiallyquality control of sera collected from lep-rosy patients needs to be identified. As iswell known, the effectiveness of blockingagents, their optimum concentrations andthe sort of microplate influences on cut-offvalues often increase and make the OD val-ues vary quite a bit among samples tested.e) it is difficult to collect the ideal sera fortesting, e.g., samples of sera from varioustypes of untreated leprosy patients and nor-mal controls from a nonendemic area of

leprosy cannot be obtained to fullill thequality requirements of TSDL.

Strategies for studying methods fordetection of il1. leprae infection. Strategiesfor studying these methodologies include:a) hefore creating a satisfactory method,one ought to stucly the methods presentlyavailable for optimization and standardiza-tion and apply them to as many availableranges as possible; b) one should stucly thepotential use of combining TSDL with thepolymerase chain rcaction; and c) one shouldcreate a new, more sensitive, more specificmethod based on serological and molecularhiology principies; an ideal confirmation testespecially needs to be developed.

SUMMARYThis is a serial stucly. In Chis series we

have estahlished 12 methods for the earlyserological diagnosis of leprosy, includingthe FLA-ABS test, ELISAs with artificialproducts (ND-O-, ND-P-, NT-O-, NT-P-BSA; PGL-I, whole M. leprae and M.sntegmatis), monoclonal antibody specificbinding assay (McAb/SBA), latex aggluti-nation test (LAT), and MLPA. These meth-ods were compared with each other on alarge scale in leprsoy patients and in thefield. The results indicate that 1) Excellentresults were obtained when ELISAs wereconducted with skim milk or egg albuminas the blocking agem and by using bloodfrom earlobes instead of from venipuncture.2) Accordng to the four "S" standard (sen-sitivity, specificity, simplicity and speed),among the 12 methods the ND-O-BSA-ELISA (ND-ELISA) is the best and theMLPA is more suitable for use in the fieldbecause it is simple and rapid. 3) In the ND-ELISA, the increase or decrease of the ODvalue has a positive correlation with the BI,and the order of positive rates was a) in var-ious types of leprosy: LL>BL>BB>BT>TT; b) in household contacts (HC), randompopulation (RP), normal controls in en-demic arcas (ENC) and normal controls innonendemic arcas (NNC): HC>RP>ENC>NNC. 4) In a population with sub-clinicai M. leprae infection, the highest riskgroup was between the ages of 15 and 25and had an increase or a persistence of highOD values prior to onset of disease. 5) ODvalues gradually decreascd over time fol-

67, 3^ Correspo ndence^ 305

lowing treatment and these declines parai-leled declines in the BI. 6) In cases curcdwith dapsone therapy, thcre was an increaseor a persistcnce of . high OD valucs in ND-ELISA prior to the onset of a leprosy re-lapse.

In conclusion, we have compared andevaluated 12 immuno-assays and haveshown that the ND-ELISA is the most prac-tical one for use in investigating sero-im-munological epidemiology, subclinical in-fection with Al. leprae, early detection ofdisease, monitoring of antimicrobial ther-apy, and even for the prediction of leprosyrelapse.

—Qinxue WuXinyu Li

Yueping YinHuiwen ShuWanhui Wei

Qi LiuGanyun Ye

Institute ofDermatologyCAMS anui PUMCNanjing 210042Jiangsu Prot'inceChina

REFERENCES1. Wu., Q.X., Li, X. Y., Wia, W. H., SHI'.N, J. P.,

SuENc, D. M. and Yo, G. Y. [Potcntial of gelatinagglutination test in the study of leprosy.] ChinaLepr. J. 9 (1993) 138-141.

2. Wu, Q. X., Li, X. Y., Y1N, Y. P. and Hou, W. [Anew SACT mcthod and its comparison with PGL-I-ELISA.] China Lepr. J. 12 (1996) 108-112.

3. Wu, Q. X., MA, Z. X., Suo, 11. W., Zuou, L. L., Liu,Q., Yi:, G. Y., WANG, T. J. and MA, B. K. [Conipar-ison of ELA-ABS test employing sera from venousblood and blood from earlobes of 79 cases of lep-rosy.] Acta Acad. Med. Sin. 7 (1985) 69-71.

4. Wu, Q. X., WI:I, W. 1-1., Liu, Q., Li, X. Y., XiI:, Z.Y. and Yu, G. Y. Ms-ELISA for detection of serumantihody leve) in patients—establishment of Ms-ELISA.] Acta Acad. Med. Sin. 10 (1988)451-455.

5. Wo, Q. X., Xue, Z. Y., Li, X. Y., Woi , W. H. andYi:, G. Y. [Detection of specitic circulating im-mune complexes in leprosy patients by mousemonoclonal antihody against phenolic glycolipid-1.) Acta Acad. Med. Sin. 11 (1989) 293-297.

6. Wu, Q. X., Yi:, G. Y. and Li, X. Y. Serological ac-tivity of natural disaccharide octyl bovine serum al-humin (ND-O-BSA) in sera from patients with lep-rosy, tuberculosis and normal controls. Int. J. Lepr.56 (1988) 50-55.

7. Wu, Q. X., Yo_, G. Y., Li, X. Y., Liu, Q. and Zuou,L. L. A prcliminary study on serological activity ofa phenolic glycolipid from Al. leprae in sera frompatients with leprosy, tuberculosis and normal con-trols. Lepr. Rev. 57 (1986) 129-136.

8. Wu, Q. X., Yi., G. Y., YIN, Y. P., Li, X. Y., Liu, Q.and Wia, W.1-1. Rapid serodiagnosis for leprosy-a prcliminary study on the latex agglutination test.Int. J. Lepr. 58 (1990) 328-333.

9. Wu, Q. X., Yi:, G. Y., Znou, L. L., Suo, 11. W., Liu,X. Y., MA, Z. X. and Li, Z. W. Determination ofantibodies in dried blood from earlobes of leprosypatients by cnzyme-linkcd immunosorhent as-says—a prcliminary report. Int. J. Lepr. 53 (1985)565-570.

Vaccination of Mice Against the Leprosy Bacillus withSkin-Test Antigens

To 'mi: EDrroi:

The specter of approximately 700,000new leprosy cases a year, despite a preva-lence of only about 805,000 (" ' 2), hauntsus. Many, perhaps the majority, of thesecases are detected in the course of intensivecase finding ("), and are single-lesion caseswhich are difffcult to diagnose definitively.Measurement of the true incidence of lep-rosy requires new diagnostic tools, more

sensitive than physical examination sup-ported at times by slit-skin smear (''). Nei-ther serology, even the modern, user-friendly, sensitive phenolic glycolipid-I(PGL-I)-based kits (S), nor gene-amplifica-tion based on a variety of primers providesthe necessary sensitivity ( 10). A promisingapproach to the diagnosis of asymptomaticleprosy is provided by the availability oftwo new skin-test antigens—MLSA-LAM(soluble antigens of Mycobacteritun leprae