In the first post, of this three-part series, we have advocated for the relevance of smear microscopy as an important diagnostic tool in the fight against tuberculosis by adding innovative capabilities that leverage the use of digital technologies. In the second post we described how TBDx™ automates smear microscopy using digital image analysis and computer-vision to deliver a 60% sensitivity improvement. In this third post we will present a diagnostic rationale for combining the strengths of TBDx™ technology with a molecular diagnostic technology such as GeneXpert (GXP) to deliver improved detection and substantial cost savings.
TBDx™ and GXP: Achieving Molecular Testing Results at Smear Microscopy Costs
As smear microscopy is practiced today, though its use, cost and performance capabilities are well understood and routinely affordable, it under-performs in identifying TB positive cases (sensitivity). In recent years more molecular diagnostic technology products have emerged, and they are making an important contribution to disease detection. For example, evaluations of GXP have presented evidence showing a 60%-70% improvement in case findings compared to smear microscopy. However, in spite of deep price subsidies, scale-up of this technology has been slowed in many countries due to operational factors and fiscally constrained laboratory budgets. Unfortunately many National Tuberculosis Program (NTP) budgets simply cannot afford to evaluate every TB specimen using GXP, particularly when 90%-95% of these cases are TB negative.
Although automating smear microscopy has been a long-held diagnostic interest within the TB research community, until recently there has been no successful launch of a commercial product with evidence of encouraging results. The emergence of TBDx™ introduces a new diagnostic technology consideration.
Today, TBDx™ holds the potential to screen TB specimens at costs similar to or lower than smear microscopy, while simultaneously achieving disease detection rates nearing that of molecular tests. As a more affordable diagnostic test, TBDx™ can screen-out a large percentage of TB negative cases at 25% the cost of a molecular test. The most likely TBDx™ positive cases then can be referred for reflex testing by GXP for both confirmation and drug susceptibility testing. This new layering of diagnostic technologies not only can fit more affordably within a laboratory budget, but also it may lead to a faster and broader deployment of molecular tests, with a more focused use. Likely more TB burdened countries would be willing to deploy molecular testing that is focused on confirming these TBDx™ positive diagnostic assessments.
Followers of our blog have read previous posts (November, 2012, April, 2012 and March, 2012) on TBDx™ and its “fit” into the world of TB diagnostics technologies. It is our view, and increasingly more the view of industry practitioners who have seen internal test performance results, that sensitivity performance and pricing makes TBDx™ an attractive screening test when combined with a more specific molecular test that can remove ‘false positive’ cases.
The performance and cost-effectiveness of TBDx™ brings into play a long-held desire by NTPs to aggressively pursue the detection of TB, where it originates and propagates, in the poorest areas – remote farming and mining communities, slums and migrant worker camps. To break the transmission cycle of the disease requires a proactive screening program, not the current reactive screening protocols. With the high-volume processing capacity of TBDx™, its portability, and ease-of-use NTPs have the platform for performing routine, recurring mass screening programs. Such a program is not feasible either with current smear microscopy (insufficient sensitivity, process too slow and not enough trained personnel) or with molecular testing (high cost per test, need for highly trained lab technicians, and controlled environmental conditions).
The combination of TBDx™ and GXP (other molecular tests could be substituted) can be brought into sharper focus in the spreadsheet (TBDx GXP Cost Analysis) that accompanies this post.
A quick look at the spreadsheet will illustrate the potential for these combined technologies:
- Achieving 85% Sensitivity / 99% Specificity;
- Reducing Cost-Per-Positive case from $118.00 to $41.00, a savings of 65%;
- Delivering a 60% improvement over smear microscopy in positive disease detection, and simultaneously lowering the Cost-per-Positive case by 30%, from $58.00 to $41.00;
- Delivering cost savings of $76.00 per-positive-case detected over the use of GXP by itself.
The combination of TBDx™ and GXP leverages the strength of each technology. TBDx™ is capable of reliably screening out negative cases, and identifying the most likely positive cases, at a substantially lower cost. GXP is capable of reliably confirming positive cases, removing those cases that are not positive, and testing true positive cases for drug susceptibility.
The benefits of this laboratory algorithm can be substantial:
- High performance at a lower cost;
- Faster and very focused deployment of molecular tests;
- Rapid diagnosis and faster patient treatment therapies;
- Lower disease transmission rates over time.
Computer-vision technologies can bring smear microscopy into the digital world we live in today and make it more relevant than ever before. Automated smear microscopy can perform alongside the best that molecular diagnostic testing has to offer, and play a vital role in helping laboratories stretch a very strained budget while making important contributions to the 2015 Millennium Goals for Tuberculosis.