The other face of Janus.

Abstract

Adalberto LaszloIn this issue, M.H.Hazbon (1) reviews recentadvances in molecular methods for early diagnosisof tuberculosis and drug-resistant tuberculosis.This article has much in common with other “stateof the art” reviews that have been appearing inthe world’s literature on molecular tuberculosis(TB) diagnostics for some time now. Most of thesereviews are well written, factual accounts ofadvances in this important field of research. Inthe last two decades much has been stated andeven more has been promised about the role ofgenomics in providing alternatives to theconventional methods that have been used forover a century.Emphasis is often given to the development ofnew diagnostic tools with turn-around timesmeasured in hours rather than days, weeks ormonths. However, in practical terms, very little ofwhat is often described with such vigorousenthusiasm offers actual relief to the modern TBlaboratory specialists or indeed much of achallenge to the old, conventional, smearmicroscopy and drug susceptibility testing (DST).As the author of the above article laments, onlytwo of these methods, based on nucleic acidamplification and detection of specificmycobacterial sequences and mutations, havebeen approved (2,3) by the Food and DrugAdministration (FDA) in the USA.Moreover, the approval for the test was given onlyfor AFB smear-positive respiratory specimens inthe case of the Amplicor test, and for either smearpositive or negative respiratory specimens in thecase of the Enhanced Mycobacteriumtuberculosis test (E-MTD). This is far from thestellar performance that the sophistication of thetechnologies used could have allowed us to expectand indeed demand. However, the recentsequencing of M. tuberculosis’s genome willeventually lead to the deciphering of the meaningof much genetic information and the creation oftruly novel TB diagnostics.The status of the replacements for the othermainstay of TB diagnostics, i.e. drug susceptibilitytesting, is even less advanced. Not only there hasnot been regulatory approval for the new tests butit has become evident over time “that the presenceor absence of mutations in drug resistanceassociated genes does not necessarily indicatesusceptibility or resistance to the correspondingdrug” (1). Isolates with mutations in DR-associatedgenes can show phenotypic susceptibility andviceversa. The so-called “real resistance-mutations” can be clinically relevant, but theirabsence doesn’t necessarily mean drugsusceptibility. These circumstances mean that testresults obtained with these methods must alwaysbe confirmed by results obtain phenotypicmethods (4-6).The existence of multiple mutations conferring re-sistance to a given drug have dampened earlyenthusiasm for these methods. Even DNAsequencing, the “gold standard” for the detectionof drug mutations, requires several sequencingreactions per isolate, and is unlikely to be usedroutinely, except for rifampicin. Otherdisadvantages include high costs, the need forsophisticated equipment and expertise, thetechnical difficulty in the performance of this testsand frequent contamination problems. Had someof these tests been developed half a century ago,physicians would have been clamouring for a more