This Expert Panel is Archived.

While this Expert Panel is no longer active, we invite you to review and recommend past replies and resources. Membership for this Expert Panel is closed, but we hope you'll join us in one of the many communities on GHDonline.

Panelists of Patient Preferences for New TB Diagnostic Tests and GHDonline staff

Patient Preferences for New TB Diagnostic Tests

Posted: 18 May, 2015   Recommendations: 9   Replies: 64

The World Health Organization is initiating the development of policy guidance on the use of the lateral flow urine lipoarabinomannan (LAM) assay for active tuberculosis in people living with HIV and will review the updated evidence for the use of Loop-mediated Isothermal Amplification (TB-LAMP) assay for detection of tuberculosis at an Expert Group meeting being convened by WHO in Geneva 1-2 June 2015. Systematic reviews are currently being performed by contracted external groups to investigate the research questions. The proposed guidance would undergo the WHO GRADE process (http://www.gradeworkinggroup.org/) a systematic process for evaluation of evidence.

As part of the process, we are interested in learning about the values and preferences of patients. Without considering the associated values and preferences, assessing large versus small magnitude of effects may be misleading. Balancing the magnitude of desirable and undesirable outcomes requires considering weighing the importance of those outcomes that is determined by values and preferences. Ideally, to inform estimates of typical patient values and preferences, systematic reviews of relevant studies of patient values and preferences should be conducted. However, there is often a paucity of empirical examinations of patients’ values and preferences.

We are therefore convening an online panel discussion to get broad input on the values of people who may potentially benefit from these new diagnostic tests for TB. Joining us for this Expert Panel are:

Dr Chris Gilpin, WHO Global Tuberculosis Programme, Switzerland
Dr Nora Engel, Assistant Professor Global Health, Maastricht University, Netherlands
Dr Karen Steingart, Cochrane Infectious Diseases Group, Liverpool School of Tropical Medicine. Liverpool, UK

Overview of LAM
Tests based on the detection of mycobacterial lipoarabinomannan (LAM) antigen in urine have emerged as potential point-of-care tests for tuberculosis (TB). Owing to suboptimal sensitivity, the urinary LAM assays are unsuitable as general screening tests for TB. However, unlike traditional diagnostic methods, they demonstrate improved sensitivity in HIV-TB co-infection which further increases with low CD4 counts. The urine lipoarabinomannan (LAM) strip-test (Determine®-TB Alere, USA) potentially can be used as a rule-in test for TB in patients with advanced HIV-induced immunosuppression, and facilitate the early initiation of antituberculous treatment in them.

Questions for consideration:
• What would be the ideal test for the diagnosis of TB among persons living with HIV?
• What level of false positive or false negative results would be tolerated?
• Would patient's want to use a test that could potentially have many false positives or false negative results?
• Would patients prefer empiric TB treatment to be initiated regardless of the LAM result?

Overview of TB-LAMP
Loop-mediated Isothermal Amplification (TB-LAMP), a unique temperature-independent way of DNA amplification, and is facilitated by a visual optic readout in an instrument that is robust and can be used at the peripheral health centre level. The TB-LAMP technology was initially evaluated by WHO in May 2013. The assay has potential as a rapid TB diagnostic tool, but the body of evidence presented on the TB-LAMP assay was insufficient during that evaluation to make a recommendation either in favor of, or against the use of TB-LAMP as a replacement test for sputum smear microscopy. Major concerns identified were for the specificity of the TB-LAMP assay particularly in low prevalence settings as well as the need for further studies in different geographical regions and in high HIV prevalence settings.

Questions for consideration:
• What would be the ideal test for the diagnosis of TB among persons with or without HIV infection?
• Would patients want a test that is more sensitive than microscopy but more costly?
• Would patient's want to use a test that could potentially have many false positives or false negative results?
• What level of false positive or false negative results would be tolerated?
• Would patients prefer empiric TB treatment to be initiated regardless of the TB-LAMP result?

The GRADE tables showing the number of false positive and negative results in different prevalence would help to make these judgments.

We are eager to hear input from all members of the community on these important issues during the Expert Panel next week, May 25 - 29. Please feel free to share questions and comments for the panel today!

Replies

 

Getachew Desalegn Replied at 9:49 AM, 18 May 2015

What would be the concern regarding stigma from co-infected patients? could
that be an issue from the patient perspective? how to address...


*Getachew Wondimagegn Desalegn (MD, MPH)*

Kedir Abdella Abdulsemed Replied at 11:29 AM, 18 May 2015

Ideal issue: The ideal TB test is the one rapid, accurate, timely, identify rapid LTBI from active, LTBI ready to progress to active TB, dead MTBC strains from live MTBC strains .....
Do patients can be benefited from LAM or TB-LAMP from these regards?
scientific concerns: I suggest LAM is serological test and urine is preferable specimen. But does kidney excretion system is the only route of excretion? or are there any other possible specimen?
what are biological underlining conduction those enhance LAM expression and release in to body fluid? or does this antigen express under any circumstance( active MTBC replication versus dormant MTBC stage )?
Do these two tests have any value in DR-TB diagnosis? or how can be they replace the recent molecular techniques those have addition value in terms DR-TB diagnosis?
Application concerns: in resource limited setting patients will need rapid tests with optimal cost with out compromising quality of results. Thus LAM and TB-LAMP shall evaluated in this regards.
Tests should have scientifically acceptable false positivety or negativity result in the manner that benefit should by over weight than disadvantage
As fur as the test not differentiate actively replicating MTBC strains from dormant intracellular bacilli , dead bacilli from viable bacilli patients specially HIV patients in resource limiting setting may clinical based empirical treatment or a sort of prophylaxis .

Kedir Abdella Abdulsemed (BSc, MSc)

Head, Mycobacteriology research center of Jimma University, Jimma Ethiopia.

Stephen Muleshe Replied at 12:50 PM, 18 May 2015

Dear Chris
Thank you very much for the invitation. Indeed I will be very glad to join
in the discussion and contribute to this new body of knowledge. Rapid
diagnostic tests for TB remain a big challenge and it will be interesting
to learn the current new developments

Thank you

MARIA ALICE TELLES Replied at 4:58 PM, 18 May 2015

Dear Chris

Thank you very much for the invitation. I will be very glad to join
in the discussion.

Urvashi Singh Replied at 9:41 PM, 18 May 2015

Dear Dr Gilpin,
Newer diagnostics are an area for discussion, development and evidence
building.It would be my pleasure to participate.
Thanks and Best,
Urvashi Singh

Valeriu Crudu Replied at 11:05 PM, 18 May 2015

Dear Chris,
Thanks for invitation.
I will participate with pleasure in this panel discussion.

Sofia Alexandru Replied at 12:34 AM, 19 May 2015

Dear Chris,
Thank you for organizing this panel!
I will be glad to participate in discussion.

DR.POOJA DALWADI GUMASTE Replied at 12:52 AM, 19 May 2015

Thank you very much for the invitation and i am excited to be a part of this panel discussion.
Its indeed my pleasure to participate.

Carolyn Tauro Replied at 1:09 AM, 19 May 2015

Dear Dr Gilpin,

Thank you for the invitation. I will be happy to join in the discussion.

Carolyn

daniel chemane Replied at 1:23 AM, 19 May 2015

Dear Cris,
Thank you for the invitation... I will be thankful for give my
contribution...

Omotayo Salau Replied at 1:27 AM, 19 May 2015

Great! Many thanks for this invitation, I am highly interested in participating.

Kedir Abdella Abdulsemed Replied at 1:37 AM, 19 May 2015

Dear Cris,
Thank for your information. I will be participate in discussion to
contribute my share.


yours sincerely,

--
Kedir Abdella Abdulsemed (MSc )

Somsak Rienthong Replied at 2:01 AM, 19 May 2015

Dear Chris,Thank you for your invitation. It's my pleasure to participate this panel discussion.Kind regards,Somsak

Marina Shulgina Replied at 2:06 AM, 19 May 2015

Dear Chris,
Thank you for the invitation. This is a very useful discussion. I think it would be helpful to intrerview groups of physicians, TB patients, other population in different countries.
Vfrina

Maopa Lewabeci Raikabula Replied at 2:20 AM, 19 May 2015

Dear Dr. Gilpin,
This is awesomely interesting. Thanks for the invitation. Will be happy to contribute a piece of my mind.

kind regards.

jayanth devasundaram Replied at 2:50 AM, 19 May 2015

All, indirectly (or directly ) related to this topic is one of innovations in Diagnostics for TB control- A USAID/Gates Foundation Grand Challenges funding opportunity if you think you have solutions.. Please go to the website and register/download the Program Overview and apply!!! http://www.gc-tbc.com/

best,

Jayanth Devasundaram

Attached resource:

Dr. Daniela Cirillo, MD, PhD Replied at 5:08 AM, 19 May 2015

Happy to partecipate
Bw

Francesco Aloi Replied at 5:13 AM, 19 May 2015

Dear Chris,
thanks for the invitation
Francesco

Regina Bhebhe Replied at 9:34 AM, 19 May 2015

Dear Chris Gilpin
Thank you so much for the invitation I am very much delighted to
participate and contributing to the panel. Regina Bhebhe PMLSc,
FIPC

Charles Sandy Replied at 9:49 AM, 19 May 2015

I will be very happy to participate in the discussions

Martin Matu Replied at 12:53 PM, 19 May 2015

Thanks Chris for the invitation.

I will be happy to join the forum and contribute to the discussion.

Best regards

Martin

Armando Matute Replied at 4:07 PM, 19 May 2015

Thank you so much for your kind invitation, it will be a pleasure to be part
of the discussion in such an interestingly TBC diagnostic topic.

Armando.

Regina Bhebhe Replied at 7:34 PM, 19 May 2015

I am delighted to participate

Ma. Tarcela Gler Replied at 9:58 PM, 19 May 2015

Thank you for the kind invite. Will be glad to join the discussion.

Tarak Shah Replied at 10:33 PM, 19 May 2015

Dear Dr Gilpin
I will be delighted to join the panel discussion
Regards Tarak

Dr Tarak Shah
Medical Officer, FIND India

Ancha Pele Replied at 12:38 AM, 20 May 2015

Thank you for the invitation... I will be happy to participate.

Samuel Kasozi Replied at 1:19 AM, 20 May 2015

Dear Gilpin,

I am happy to participate in the panel discussions.

Regards

Dr.Samuel Kasozi.
TB Survey Coordinator, School of Public Health, Makerere University

Yasuhiro Yasutomi Replied at 2:37 AM, 20 May 2015

Dear Dr.Glipin,

Thank you very much for invitation to such great meeting. I am happy to join in the discussion.
The TB-LAMP is now widely-distributed diagnostic test for TB in Japan. Most of hospital in area of high incidence of TB infection used this system. Introducing and running costs of TB-LAMP are much lower than Gene-Expert.
The Ag detection system for Mtb such as TB-LAMP or Gene Expert is not affected by host immune system unlike in the case of IGRA, so these systems are useful in immunocompromised person such as HIV infection.

Kind regards,

Yasuhiro Yasutomi

Mark Harrington Replied at 2:43 AM, 20 May 2015

Dear M. Yasu:

LAMP and GX are NAAT- reading systems -- neither Ab nor Ag!

Yasuhiro Yasutomi Replied at 2:55 AM, 20 May 2015

Sorry, my explanation was incorrect. TB-LAMP is not Ag detection sysem (gene detection).

Yasuhiro Yasutomi

SYDNEY CHIKUKWA Replied at 3:11 AM, 20 May 2015

May someone advise on the costs of TB-LAMP system including equipment and
where can one source the instrument. Does the instrument also have the
capacity to handle other tests as gene expect which does chlamydia and
gonorrhoea as well as HIV viral load

Mark Harrington Replied at 3:16 AM, 20 May 2015

But more importantly, we need TB tests to guide treatment -- and we need you Yasu!

Ruth MCNERNEY Replied at 3:43 AM, 20 May 2015

Two technical clarifications.

1. LAMP is described as temperature independent. I don't believe this is true, the reaction takes place at 65C and needs a heating block or oven at a steady temperature. It should be noted that two other commercial rapid NAAT tests are now available and are being evaluated, one from India http://www.bigteclabs.com/newsroom-launchday.html and one from China http://www.bioustar.com/en/ We wait to see how accurate, robust and easy to use they are.

2. The LAM antigen detection urine test described is a commercial product. The antibodies were first used in an ELISA format and then a lateral flow device. The sensitivity achieved was low and the specificity of the antibodies used is not great. A publication from Tanzania suggests false positives may be caused by contamination of samples by dust or faeces. There are several ongoing efforts to develop an improved test either by using better antibodies or better detection technology and a recent publication from a Swedish group suggests increased sensitivity is achievable. Please don't write off the possibility that LAM detection might one day be a useful test for screening all patients because the first commercial product to be sold was less than optimal.

kind regards
Ruth

Haluk Çalışır Replied at 4:44 AM, 20 May 2015

Dear Dr Gilpin,
Thank you very much for your invitation. I would love to join the
discussion.
Best REhards,
Haluk Çalışır M.D
Assoc. Prof.
Pulmonology
Acibadem University
Istanbul/Turkey

Balasangameshwara Vollepore Replied at 5:32 AM, 20 May 2015

Dear Dr Gilpin,

Thank you and I accept your invitation.

Panganai Dhliwayo Replied at 9:07 AM, 20 May 2015

Thank you for the invitation. I will participate.

Shubhada Shenai Replied at 10:24 AM, 20 May 2015

Thank you Dr. Gilpin! I would like to join this discussion.

VIDYANIDHI GUMMA Replied at 11:39 AM, 20 May 2015

Dear Dr. Gilipin

Many thanks for invitation, I am happy to join in the group.

Florian Sauter Replied at 1:47 PM, 20 May 2015

Dear Dr. Gilipin,

thank you for the invitation! I will be more than happy to join the discussion!

NOWSHAD ALAM Replied at 3:37 PM, 20 May 2015

DEAR DR. GILPIN,

THANK YOU FOR THE INVITATION. I AM VERY MUCH INTERESTED TO JOIN IN THIS DISCUSSION.
IT IS VERY IMPORTENT TO CLEARIFY THROUGH DISCUSSION, SPECIALLY PRIOR TO PUT IN ROUTINE PROTOCOLE FOR THE INVESTIGATION OF TB, PARTICULARLY HIGH BURDEN OF HIV ASSOCIATED TB.

REGARDS.


DR. ALAM
PORT ELIZABETH, SOUTH AFRICA

Juma Khudonazarov Replied at 3:47 PM, 20 May 2015

Hi Christopher,

Thanks more than happy to discuss the proposal. If indeed (LAM)can be used
as a rule-in test for TB in patients with advanced HIV-induced
immunosuppression that would be huge progress. I have few points to rise,
but overall it can be another breakthrough and help both patients as well
as doctors. Looking forward to discussion.

Ntihabose Corneille Killy Replied at 3:47 PM, 20 May 2015

I am looking forward starting of this discussion.

Indrajit Kumar Replied at 12:59 PM, 22 May 2015

Respected sir,
Thank you for the invitation. I will be happy to join in the discussion.
Regards
Indrajit.

Fikreselam Desalegn Replied at 1:50 PM, 22 May 2015

Dear Dr. Christopher Gilpin, Thank you very much for inviting me for this important meeting. I am here for the discussion.
A- The ideal diagnostic test would be the one most accurate, least invasive, rapid, affordable, simple to handle and transport and which can be used in the field ( at the point of care of patients). Until now there is no such TB diagnostic test that fulfills all of the criterias. To my knowledge The Xpert MTB/RIF test is the most innovative one, better to fulfill the requirements of “the ideal TB diagnostic methods” with the additional benefit of detecting rifampicin resistance in a TB population. Companies and researchers are still striving to develop the ideal TB diagnostic test which is more perfect and simplified than the Xpert MTB/RIF system which can be used in all hospitals, health centers and health posts. From these new developments are the urinary LAM test and LAMP tests. These discoveries are another breakthrough in the diagnosis of tuberculosis and the innovators are really be thanked and rewarded for their efforts. Every TB diagnostic test might have some drawbacks when recommended programmatically but to the individual patient, they are all important alternative diagnostic means. Therefore their presence in the market is very crucial both to the patient and the clinician. As evidence accumulates from their use in the future, and as long as the innovators make improvements from the feedback, sure we will have the ideal diagnostic test for this deadly disease!
I tried to see some of the articles about these two tests- the urinary LAM and LAMP and want to raise points: The Urinary LAM test detects the component of cell wall, Lipoarabinomannan (LAM) antigen, which is filtered by the kidneys to appear in the urine. But how is the degree of shading LAM in the urine? Is that all uniform in all patients? Does it affected by the degree of renal function? Does the presence or absence of some molecules which are filtered at the glomerules affect its excretion? LAM is a component of all mycobacteria species so how can we be sure that it is not from NTM in the field (at the point of care of patients)? . TB-LAMP: has the advantage of being self contained, not requiring high biosafety level, not simplified? however its TB detection in smear negatives is not more than 60%. What about diagnosis of extrapulmonary TB ? What about in HIV infected and other immunosuppressed individuals? also the different studies have different out comes, further studies? cost and training issues? Can these two tests be modified in some way to have a roll in drug resistant TB?
B- What level of false positive or false negative results would be tolerated? I guess, On one thing we all agree-A NEW DIAGNOSTIC TEST TB DETECTION CAPACITY SHOULD BE GREATER THAN SPUTUM SMEAR MICROSCOPY TO BE RECOMMENDED FOR ROUTINE PATIENT SCREENING!! Having said this, a new TB test should have:
1. In smear positive and culture positive cases a sensitivity and specificity of greater than 95% compared to culture
2. In smear negative and culture positive cases a new test should have a sensitivity at least 75% compare to culture and specificity 95%
3. In extrapulmonary cases regardless of the site – over all sensitivity of at least 75% and specificity greater than 90% compared to culture(pathology may help for clinical decision)
4. In advanced HIV infection(CD4< 200/µl) and in under five children a sensitivity and specificity of at least 80% ( difficult patients)
These are not from studies just my views to have acceptable values, made from experience of the recommended TB diagnostic modalities( smear, GeneXpert, LPA, conventional culture and DST..) and whether alternative means are present for clinical decision.

C- Would patients want to use a test that could potentially have many false positives or false negative results? Considering the deadly nature of the disease and it is curability, in high burden countries and HIV prevalent areas , I think patients would prefer a test that has more false positives than false negative result (not for a reason!!). In low prevalent areas patients may prefer a test which has low false positives and low false negative results.

D- Would patients prefer empiric TB treatment to be initiated regardless of the LAM result? In my country clinicians and patients don’t have experience with this test, but from the studies, it has promising result in patients with advanced HIV infection. So I think further parallel studies are needed before reaching a conclusion. From the patients’ perspective, patients need evidence for their treatment.
Thankyou !!

Celso Khosa Replied at 3:56 PM, 23 May 2015

Dear Dr. Gilpin;

Thank you. I'm looking forward to my participation in the discussion.

Gunturu Revathi Replied at 9:42 AM, 25 May 2015

Dear Chris Gilpin,
Thank you very much and I am certainly interested. We did a small study of evaluation of LAMP test in our TB patients and would like to discuss thee issues.

Karen Steingart Replied at 11:44 AM, 25 May 2015

Greetings. I wanted to reply to the questions about whether the urine LAM test can rapidly distinguish latent TB infection from active TB disease and detect drug resistance. The urine LAM test detects active TB (pulmonary and extrapulmonary TB) and is not a test for drug resistance.

As many of you know, the LAM test we are discussing is the urine lateral flow lipoarabinomannan assay. This is a commercially available point-of care test for active TB disease (Alere DetermineTM TB LAM Ag, Alere Inc, Waltham, MA, USA). The test detects lipoarabinomannan (LAM), a lipopolysaccharide present in mycobacterial cell walls, which is released from metabolically active or degenerating bacterial cells and appears to be present only in people with active TB disease. LAM is detectable in the urine of people with active TB. A meta-analysis of an earlier generation LAM-ELISA test (Minion 2011) found that the accuracy of LAM-ELISA for TB was higher in people with HIV infection (than in those without HIV infection), especially in people with advanced immunosuppression. Several hypotheses may explain the higher sensitivity of urinary LAM detection in people with HIV including higher bacillary burden and antigen load, greater likelihood of genitourinary tract TB involvement, and greater glomerular permeability to allow increased antigen levels in urine. (See also Cox J, Is Urinary Lipoarabinomannan the Result of Renal Tuberculosis? Assessment of the Renal Histology in an Autopsy Cohort of Ugandan HIV-Infected Adults. PLoS ONE 10(4): e0123323. doi:10.1371/journal. pone.0123323.)

So, right now, with the commercially available urine lateral flow LAM assay, the interest has been on the accuracy of the test for detection of active TB in people with HIV, people who suffer high morbidity and mortality. If accurate, the test could provide obvious benefits, by earlier detection of pulmonary TB that may be missed by sputum microscopy and sputum Xpert® MTB/RIF and extrapulmonary TB that may be missed by sputum-based testing. …Karen Steingart, MD, MPH

Karen Steingart Replied at 12:03 PM, 25 May 2015

Thank you for the question about nontuberculous mycobacteria and LAM. A recent paper addressed this question: Qvist et al. Urine lipoarabinomannan point-of-care testing in patients affected by pulmonary nontuberculous mycobacteria – experiences from the Danish Cystic Fibrosis cohort study. BMC Infectious Diseases 2014, 14:655
http://www.biomedcentral.com/1471-2334/14/655. The authors concluded.“This is the first study to assess urine LAM detection in patients with confirmed NTM infection. The study demonstrated low cross-reactivity due to NTM infection when using the recommended grade 2 cut-point as positivity threshold. This is reassuring in regards to interpretation of the LAM test for TB diagnosis in a TB prevalent setting. The test was not found suitable for NTM detection among patients with CF.”

Rajbir Singh Replied at 12:00 AM, 26 May 2015

Are contents or summary of discussion available online?
Yesterday could not join the session.
Regards,

Nora Engel Replied at 4:06 AM, 26 May 2015

Dear All,

I wanted to pick up an earlier comment on whether patients would prefer false positive or negative TB test results.
Fikreselam Desalegn wrote: "C- Would patients want to use a test that could potentially have many false positives or false negative results? Considering the deadly nature of the disease and it is curability, in high burden countries and HIV prevalent areas , I think patients would prefer a test that has more false positives than false negative result (not for a reason!!). In low prevalent areas patients may prefer a test which has low false positives and low false negative results."

- Dear Fikreselam Desalegn, could you expand on why you think patients would prefer false positives over false negatives?
- Do others agree with this view?

- And a more general question for the patients among the discussants:
Did you ever receive a wrong test result (for HIV/TB status, false pos/neg)? Or were there ever uncertainties about your test result not being clear/correct (discordant results for instance)? Why/ what happened?

- And a more genreal question for the providers:
What are your experiences with discordant/false results? How did patients deal with those? What were challenges?

Thank you for your answers!
Nora Engel

Kedir Abdella Abdulsemed Replied at 4:52 AM, 26 May 2015

Dear all i would like rise issue of LAM and LAMP.

The below are some questions i replayed early in the discussion. But i thought that may be before the time of discussion and some can react on those questions now. In addition i would like to rise some of essential questions as follows.
Ideal issue: The ideal TB test is the one rapid, accurate, timely, identify rapid LTBI from active, LTBI ready to progress to active TB, dead MTBC strains from live MTBC strains .....
Do patients can be benefited from LAM or TB-LAMP from these regards?
scientific concerns: I suggest LAM is serological test and urine is preferable specimen. But does kidney excretion system is the only route of excretion? or are there any other possible specimen?
what are biological underlining conduction those enhance LAM expression and release in to body fluid? or does this antigen express under any circumstance( active MTBC replication versus dormant MTBC stage )?
Do these two tests have any value in DR-TB diagnosis? or how can be they replace the recent molecular techniques those have addition value in terms DR-TB diagnosis?
Application concerns: in resource limited setting patients will need rapid tests with optimal cost with out compromising quality of results. Thus LAM and TB-LAMP shall evaluated in this regards.
Tests should have scientifically acceptable false positively or negativity result in the manner that benefit should by over weight than disadvantage
As fur as the test not differentiate actively replicating MTBC strains from dormant intracellular bacilli , dead bacilli from viable bacilli patients specially HIV patients in resource limiting setting may clinical based empirical treatment or a sort of prophylaxis
1. LAM screened only for active TB and no and also depend on antigen which in fact excreted from actively replicating bacilli. But is true that MTBC is extracellular bacteria and the antigen concentration to be detected in urine sample can be influenced by the rate of bacilli replication plus reassessment of bacilli from macrophage. Can we estimate the number(load of bacilli to be present in fluid just to have detectable antigen by LAM? If that is the cases we can suggest the merit of LAM over Xpert (which can detect about 130 bacilli). The other issue is the DR-TB. Now a day Public health importance of TB is not only burden of disease of but DR-TB which in fact increases the mortality form TB. Therefore, my questions is if we diagnosis patents for active TB using LAM are we going to perform DST for the patients being positive by LAM or are going simply put on treatment.
If we have have DST for them what is the advantage of this test as compared other molecular assay?
2. Regarding patients preference of false positive and false positive test i share the fkereselem's idea look in developing country physicians con initiate TB treatment based on symptoms. But 1. TB symptoms is not specific and miss diagnosis can affect patients, patients family and even public . On top of sign and symptom including tests with false positive and false negative result but bellow 15% for each may have added value to TB diagnosis. Regarding empiric treatment this day the issue of DR-TB is one of the head ace to TB control and to me empiric treatment my have contribution for DR acquiring in MTBC. Hence, it is advisable to use all the available test alternatives before treatment initiation. This can be address via creating bridge among laboratories to access all the alternatives.

Thanks

Christopher Gilpin, PhD, MPH Replied at 5:00 AM, 26 May 2015

We would all prefer to have the ideal test for the diagnosis of TB, that is one that has a sensitivity of 100% and a specificity of 100%. In the real world there are not perfect tests. Hence we need to consider the benefits to patients when using a particular test as well as considering and any adverse effects for the the patient when using a particular test.
If a test has low sensitivity, it means that any patients with TB (true positives) would go undiagnosed and may not be put on proper treatment. If a test has low specificity, it means that patients may be wrongly diagnosed with TB (false positives) and be put on TB treatment unnecessarily.
In the population being tested we need to balance the performance of the test considering the relative numbers of true results with a test with the relative numbers of false or incorrect results.
One diagnostic technologies being evaluated by WHO using a Guideline Development Group – urine lateral flow lipoarabinomannin (LF-LAM) assay for the diagnosis of TB in persons with HIV. Given the difficulty of diagnosis TB in PLHIV – what should be the minimum performance of a test to be acceptable to patients? Is it acceptable for a test to yield the same number of true positive results as false positive results? Should the proportion of true positive results be much greater than the proportion of false positive results? If so, how much greater? What if the test generates more false positive results than true positive results?
Patients should have evidence for their diagnosis of TB before starting TB treatment, however, when HIV positive patients are seriously ill with symptoms of TB should treatment for TB be started regardless of a test that gives false positive and false negative results. In seriously ill persons with symptoms of TB when would patients want empiric treatment started regardless of the test result or no test at all?

jayanth devasundaram Replied at 5:14 AM, 26 May 2015

Not sure why everyone only thinks of one test - a two stage screening test would address the specificity/sensitivity limitation of one stage tests.


JD

Fuad Mirzayev Replied at 5:47 AM, 26 May 2015

on comment from jayanth devasundaram
Tests for screening and then second stage diagnostic tests should also have certain characteristics both in terms of performance and the cost. Screening test should have high sensitivity and ideally should not be too costly as it is supposed to be used for a large numbers of presumptive cases. Second stage diagnostic tests should be highly specific and will need to be done on all "positives" of the first-screening test. Cost of case identified will add up cost of the diagnostic test and all the "positive" screening tests of cases that were confirmed and not confirmed by the diagnostic test.
Another possibility when the current practice testing is not ideal (not sensitive but specific) is to use an add-on test to test all "negatives"

Fikreselam Desalegn Replied at 1:51 PM, 26 May 2015

dear Nora Engel Thank you for raising the issue:
The ideal TB test is as I stated in my previous discussion, but lack of that and difficult to diagnose TB in some cases, a test with high sensitivity compared to specificity is better to make a decision together with clinical evaluation in critically ill patients in high burden areas and in people with immunosuppression. The patient also will have a reason to adhere to the treatment(low default rate). If the test specificity compared to the sensitivity is high, some patients are going to be treated empirically. But this will not give them enough reason to adhere to treatment and also will be difficult to the HCW to convince the patient from the outset. In low burden countries the positive result can be confirmed by a secondary test such as culture/genexpert...etc. in my clinical experience most people are aware of TB (Ethiopia is among one of the high burden countries) and if you miss TB they don't think that it is b/c of failure of the tests to diagnose TB, but it is b/c the clinician failed to diagnose and treat TB. Some patients say "the health care worker failed to diagnose my TB, then when I start ART(antiretroviral treatment) my condition worsened". This is the reality at the ground
Four things should be clear from a clinical point of view:
1. TB is a treatable condition, if patients take the drugs appropriately ,i.e, we are not diagnosing a life long illness/fatal disease with no treatment
2. Treating a tb patient means not only relieving suffering from the patient, but also protecting the community
3. It is difficult to say no TB in patients with advanced immunosuppresion by the currently available tests- most of these patients have Either active TB silently eating them or a transforming latent TB which is going to be full blown when the immunity is restored.
4- Patients need evidence for their treatment

Ellen Baron Replied at 9:02 PM, 26 May 2015

I am responding to Christopher Gilpin's request for comments concerning what level of imperfection "we" and patients should tolerate with regard to a diagnostic test, given that no test is perfect (something I emphasize constantly). I believe that results of a diagnostic test should be used to support a treatment decision. If the test results are not reliable enough to change a patient presentation-based decision by the caregiver to treat empirically (or not to treat), at least in the case of tuberculosis, then why spend the money for the test at all? Tests should be used judiciously based on a reasonable pre-test probability that the disease is present or possibly present. Clinical criteria for suspicion of TB are pretty well described in many populations. If the infrastructure to carry out the appropriate patient management is present and used, and if the patient population can be reliably characterized, then algorithms for patient management based on test results can and should be implemented. The sensitivity and specificity of the test in question must be determined in the intended patient population before deciding whether to use the test in that situation at all, and with that information, then an appropriate testing algorithm can be created. This is my own personal opinion, but do note that I work for Cepheid in my day job. Ellen Jo Baron, Prof. Emerita, Stanford Univ. Ex.Director Medical Affairs, Cepheid

Vineet Bhatia Replied at 12:38 AM, 27 May 2015

I agree with Jayanth Devasundaram. We should not get fixated with finding 'the single most ideal' test. We could have a two stage testing as well. As Fuad said the first stage could be a high sensitivity test followed by a specific test - something like the ELISA and Western Blot for HIV. Yes, on the face of it, this would mean added cost of tests but then we need to compare it with cost of excess treatment/ no treatment to the deserving. I feel the patients need the right treatment at right time with right advice which cautions them but does not scare them. Therefore the diagnostic procedures should enhance the confidence of the patient in the process that s/he would have to go through, and that the process would lead the patient to being cured.  Of course, complementing a good diagnostic process with an ideal treatment is necessary but probably the subject matter of another discussion.

Nora Engel Replied at 6:24 AM, 28 May 2015

Dear All,

One of the questions was: Would patients prefer empiric TB treatment to be initiated regardless of the LAM result?
So far it seems that some of you argue that patients would not adhere to empiric treatment as suggested by the health care worker if not supported by a test result.
- Do others agree with that?
This situation might be very different in different contexts: For instance, from our research in India it became clear that lots of times patients would prefer empiric treatment over having any tests done to avoid the extra cost and loss of time associated with testing. On the other hand, private providers often did not order tests for TB because of the stigma attached to the sputum sample (they feared losing their patients) which could be overcome by another sample.

Thus it might also be helpful to talk about the status quo of TB testing among HIV patients:

For the providers:

1. When and where are people with low CD4 count screened for TB? how and by whom? what are current problems and experiences with that?
2. How are low CD4 counts detected? How often do people who are HIV positive monitor their status and how? could TB testing be integrated into that if the sample is urine? by whom?
3. What are experiences with other urine based tests at clinics/community outreach settings/homes? what are problems with it?

For the patients:
Have you undergone empiric (preventive) TB treatment before? would you agree to it regardless of test result if your doctor suggests? why (not)?

Best,
Nora

Nora Engel Replied at 8:22 AM, 29 May 2015

Some more thoughts related to that:
some people emphasize that test results have a use for ensuring adherence to provider advice. This is something that we also found in our research on diagnostic practices at POC: Tests have different purposes in different situations, for instance ensuring treatment adherence, easing acceptance of a disease, entrusting confidence in care provided, accessing a government grant or a specific follow up test. but that is difficult to foresee for WHO and difficult to make a decision upon.

We also noticed that tests are talked about among patients and if a test is known to be unreliable (e.g. lots of false positives or negatives) then patients will take that into account in their decision to seek a diagnosis or act on a clinicians advise .

Karen Steingart Replied at 10:06 AM, 29 May 2015

Dear Everyone,

I am posting replies for Dr Stephen Lawn to previous questions

Kedir Abdulsemed asked if we diagnose patients for active TB using LAM, are we going to perform DST for the patients being positive by LAM, or are we going to simply put the patient on treatment?

Reply. Should follow WHO guidance. We would like to have DST on all patients with a TB diagnosis. So, if for example, culture of sputum or another available sample is possible, then this would be ideal. Otherwise, as with smear microscopy, treatment regimen decisions in the absence of DST may need to take into account data on local prevalence of drug resistance or the drug susceptibility of a known contact / source case.

Fikreselam Desalegn mentioned that if the test specificity compared to the sensitivity is high, some patients might be treated empirically. But this might not give patients enough reason to adhere to treatment and might be difficult for Health Care Workers to convince the patients from the outset.

Reply. I presume FD is concerned that if the sensitivity is low, the NPV will be low and therefore clinicians may give empirical TB Rx. LAM might best be used as part of a package of investigations that collectively provide higher sensitivity and NPV that LAM alone. Adherence to empirical TB Rx (started in the absence of a positive microbiological test) will be an important outcome of several strategy trials (ACTG REMEMBER; TB FAST TRACK) assessing the impact of empirical TB strategies in HIV+ individuals that are likely to report in the next 12 months or so.

Karen Steingart posted for Dr Lawn

Karen Steingart Replied at 10:08 AM, 29 May 2015

All,

Here is another reply from Dr Maunank Shah

• Kedir Abdulsemed asked if we diagnose patients for active TB using LAM, are we going to perform DST for the patients being positive by LAM, or are we going to simply put the patient on treatment?

• Fikreselam Desalegn mentioned that if the test specificity compared to the sensitivity is high, some patients might be treated empirically. But this might not give patients enough reason to adhere to treatment and might be difficult for Health Care Workers to convince the patients from the outset.

From Dr Shah:

For question 1: This testing algorithm could be considered analagous to any of our existing tests which do not provide immediate DST results--such as sputum smear microscopy or mycobacterial culture. Depending on the background prevalence of drug-resistant TB, most providers are likely to start empiric therapy with 1st line drugs (similar to receiving a diagnosis of TB on the basis of a positive sputum smear, or a positive mycobacterial culture). In areas or populations with higher prevalence of drug-resistance testing, the algorithm for diagnosis and treatment is likely to include a follow-up test to determine DST (i.e. mycobacterial culture+DST, or a NAAT that provides resistance test results).

For question 2: I am somewhat unclear on what is being asked. But in general, I think the precedence of and usage of sputum smear-microscopy is a very closely related topic (a test with poor sensitivity, and high but suboptimal specificity). Most TB programs would treat on the basis of a positive sputum smear, and patients would generally accept this result. While LF-LAM may take time for HCW to become comfortable with, it would be considered similar to acting upon a positive sputum smear result (performance characteristics are in fact extremely similar). Some programs may choose to perform additional testing with mycobacterial culture for additional laboratory confirmation (similar to algorithms using smear-microscopy).

Lal Mani ADHIKARI Replied at 7:25 AM, 2 Jun 2015

Dear Dr. Gilpin,

Thank you for your invitation to the participation. As I am out of office
until 9th June 2015 engaged in post disaster emergency Response. I could
not attend the conference, I will be grateful to get the update from you
and the other expertise.

Many thanks,

Sincerely,
Lal Mani

Isiramen Olajide Replied at 11:26 AM, 2 Jun 2015

Thanks for the invitation; TB diagnosis at resource limited settings is a challenge due to infrastructure and lack of personnel, thus a rapid and low cost/no cost test. The LAMP test will be an advantage if it can provide a rapid, low cost and quality (short of false positive and false negative result) result. One major short coming of the smear microscopy is the two or three visits, TB suspects have to make before results are released.
When the LAMP test is proven to be sensitive in TB/HIV co infected patients, its use will ensure initial increase in diagnosis. I look forward to a rapid POC quality test for TB diagnosis.

Anita qaffari.MD Replied at 4:31 PM, 4 Jul 2015

Evaluation of GeneXpert MTB/RIF for Diagnosis of Tuberculous Meningitis
Nguyen Thi Quynh Nhua, Dorothee Heemskerka, Do Dang Anh Thua, Tran Thi Hong Chaub, Nguyen Thi Hoang Maib, Ho Dang Trung Nghiab, Pham Phu Locb, Dang Thi Minh Haa,c, Laura Mersona, Tran Thi Van Thinha, Jeremy Daya, Nguyen van Vinh Chaub, Marcel Wolbersa, Jeremy Farrara and Maxine Cawsa
aOxford University Clinical Research Unit, Hospital for Tropical Diseases, Wellcome Trust Major Overseas Programme, Ho Chi Minh City, Vietnam
bHospital for Tropical Diseases, Ho Chi Minh City, Vietnam
cPham Ngoc Thach Hospital for Tuberculosis and Lung Diseases, Ho Chi Minh City, Vietnam
K. C. Carroll, Editor

This Expert Panel is Archived.

While this Expert Panel is no longer active, we invite you to review and recommend past replies and resources. Membership for this Expert Panel is closed, but we hope you'll join us in one of the many communities on GHDonline.

Panelists of Patient Preferences for New TB Diagnostic Tests and GHDonline staff