Publications

By far the most common method of TB diagnosis is acid-fast bacillus (AFB) smear microscopy, in which a patient’s sputum specimen is treated with staining reagents and examined under a microscope to detect rod-shaped bacteria or bacilli. This method is rapid, inexpensive and identifies the most contagious patients. But it has serious limitations.

Smear microscopy identifies just 45% to 80% of TB cases. It also has poor positive predictive value (a measure of confidence in a positive result), since it does not distinguish between Mycobacterium tuberculosis and nontuberculous bacilli that are also acid fast.

Because of these shortcomings, AFB smear-positive specimens and AFB smear-negative specimens from patients with known TB risk factors must undergo confirmatory testing. Culture—in which M. tuberculosis is grown in a Petri dish or culture tube—is the gold standard confirmatory test and is required to isolate bacteria for drug-susceptibility testing and for genotyping to characterize TB strains.

The problem is M.tuberculosis is a notoriously slow-growing organism. It can take two to four weeks for the mycobacteria to grow in culture, and laboratorians cannot rule out the possibility of M. tuberculosis growth until at least six weeks have elapsed. Once the organism is isolated and confirmed as M. tuberculosis, additional testing to gauge drug resistance can take up to four weeks longer. Thus, physicians and patients with drug resistant infections can wait anywhere from three to ten weeks before learning the appropriate course of treatment and definitively breaking the chain of transmission.

Said Castro, “Because you’re dealing with a slow-growing bug, you need to beat it. One way to beat it is by reading its DNA before it has time to replicate.”

New Technologies for TB

In 1995, the FDA approved the first DNA test to diagnose TB directly from sputum specimens, a nucleic acid amplification (NAA) test produced by Gen-Probe for use with AFB smear-positive respiratory specimens. Four years later, the agency approved an enhanced Gen-Probe NAA test for use with AFB smear-negative specimens
from patients suspected to have TB.

NAA testing is more reliable than smear microscopy and more rapid than culture, taking just hours to generate results. CDC has recommended its use since 1996. The most recent guidance—based on input from a panel of experts convened by APHL and CDC—was released in early 2009 and recommends NAA testing “on at least one respiratory specimen from each patient with signs and symptoms of pulmonary TB for whom a diagnosis of TB is being considered but has not yet been established and for whom the test result would alter case management or TB control activities.”

A second testing milestone was the September 2009 launch of CDC’s Molecular Detection of Drug Resistance program, a laboratory service that accepts TB isolates from public health laboratories (PHLs) and other submitters and provides preliminary susceptibility data for eight first- and second-line anti-TB drugs, based on examination of genetic mutations associated with drug resistance. Results are confirmed in parallel with culture-based susceptibility tests.

The program sidebar, pg 17, has so far, examined isolates for about 150 patients, comprising half of all multidrug-resistant TB (MDR TB) cases reported nationally during that time. It advertises a turnaround- time of four days, but often delivers results sooner. The real delay comes at the front end, since submitting laboratories must still isolate M. tuberculosis in culture in order to submit the necessary isolates.

The latest technological innovation- Cepheid’s GeneXpert MTB/RIF assay-bridges this gap. It not only provides positive or negative results for M. tuberculosis from sputum specimens within two hours, but simultaneously detects mutations associated with resistance to rifampin, one of the most important anti-TB drugs. In many settings, especially internationally, rifampin resistance is considered a surrogate marker for multidrug resistance.

The MTB/RIF assay has generated much excitement in the international public health community. Last December, the World Health Organization (WHO) endorsed the assay, saying it could “revolutionize TB care and control” and enable a three-fold increase in the diagnosis of patients with drug-resistant TB.

Promise Deferred

Where it has been instituted, molecular TB testing has been a boon to laboratories, healthcare providers and TB controllers. The Florida Department of Health Bureau of Laboratories (FBL) was an early adopter, implementing the first NAA test kit for TB within one month of FDA approval. In 2009, the Florida PHL network tested roughly 24,000 clinical specimens for the disease, and the FBL performed more NAA tests for TB than any other PHL.

Overall, though, Max Salfinger, MD, a TB expert and Florida state laboratory director, said implementation of the 15-year-old technology has been “too slow.” In fact, most US PHLs have not yet been able to fully implement CDC’s 2009 NAA test guidelines.

One reason is cost. Even though early TB diagnosis and treatment would likely generate overall savings for state TB control programs, the expense to individual laboratories is considerable. Operating costs associated with Gen-Probe’s NAA assay, for example, can exceed $100 per test (depending on testing volume), placing it beyond the reach of many PHLs at a time when most states face budget deficits.

Kelly Wroblewski, MPH, MT(ASCP), director of APHL’s infectious disease program, said PHLs are generally worried about maintaining existing services, given the current economy. Adding new technology, she said, “is difficult and involves huge training components.” [See pg. 17 sidebar for information about a one-time CDC grant that is aimed at helping labs ramp up TB NAA testing.]

But there are other barriers. Chief among these is lack of commercial availability of Cepheid’s highly touted MTB/RIF NAA assay. The assay is in use in the European Union and at least seven other countries—Australia, Brazil, Columbia, Korea, Mexico, Panama and Venezuela—but awaits US FDA approval for domestic use.

Because TB prevalence is so low in the US, and the prevalence of MDR TB even lower, the pre-market clinical evaluations required for FDA approval pose greater challenges here than in high-prevalence settings. (When a disease is rare, the major determinant of positive predictive value is disease prevalence in the screened population.)

Sally Hojvat, PhD, director of FDA’s Division of Microbiology Devices, said, “FDA realizes that, in the interest of public health, there is a definite need for such a test to be available in the US, and we will work diligently with our sister agency, the CDC, and the company to enable this to occur as quickly as possible, with the important caveat that the safety and efficacy of the test is demonstrated in the US population.”

Russel Enns, PhD, Cepheid’s chief regulatory officer, said the company anticipates product approval and availability in the US in 2013. In the meantime, CDC officials are hoping to collaborate with the FDA to secure an investigational device exemption (IDE) that would enable laboratories to use the assay with appropriate mitigation procedures and quality controls.

“We ought to assertively try to use it in the US and not just wait for the next two years of data collection for it to happen,” said Castro. “The IDE would enable the use of this promising diagnostic device while collecting additional key information to confirm safety and validate performance.”

At least one PHL, the San Francisco Department of Public Health Laboratory (SFDPHL), has performed an extensive in house validation of the MTB/RIF NAA assay and is already using it. Mark Pandori, PhD, HCLD(ABB), the lab’s director, said SFDPHL scientists tested 120 specimens of varying degrees of positivity, including specimens containing non-tubercular mycobacteria, and found the Cepheid assay to be as sensitive as the Gen-Probe NAA assay for smear negative, culture-positive specimens—“the real challenging specimens.”

The SFDPHL implemented the assay in April 2010. Last year, the laboratory tested 2,385 sputum specimens and confirmed 98 TB cases, including 17 drug-resistant cases and 3 multidrug-resistant cases.

With the MTB/RIF NAA assay, Pandori said, molecular testing is “no longer a major issue; we can turn it around in two hours or less with very little impact on the labor situation.”

Because of the tremendous difference in the time to diagnosis— two hours versus several weeks—Pandori said, “If you believe we have an honest obligation to eliminate TB, then you have to embrace molecular technology.”

Low-incidence states, however, face extra challenges. When the number of specimens submitted for testing drops below a certain threshold, it becomes increasingly difficult for laboratorians to maintain technical proficiency, and the cost per test increases.

Andrea Labik, ScD, head of the West Virginia state PHL, said her lab was “on the verge of getting rid of TB testing altogether” about a decade ago precisely because of low specimen volume. Laboratory staff and the state TB control officer embarked on an outreach campaign to “drum up business” from hospitals that had been sending their suspect specimens elsewhere for testing.

The campaign worked. Last year, the laboratory received 1,308 clinical specimens and identified 48 TB cases using Gen Probe’s molecular assay. However, specimen volume—roughly two dozen per week—is still insufficient to enable scientists to maintain proficiency in drug susceptibility testing. Thus, the West Virginia lab sends TB isolates across state lines to the Pennsylvania Bureau of Laboratories for drug-resistance testing.

Other low-incidence states have devised similar arrangements to maximize efficiencies and assure prompt, reliable testing.

Although California vies for position as the highest incidence state in the country, TB specimens are divvied up among the state PHL, three dozen or so local PHLs and a few large commercial labs. Ed Desmond, PhD, D(ABMM), who oversees mycobacterial disease testing at the state laboratory, said his staff tests about 2,000 specimens per year.

Some of those specimens are sent for testing through an innovative program devised to support California’s rural, local PHLs. Those labs, said Desmond, have highly qualified staff, but find it uneconomical to invest in the expensive instrumentation needed to keep up with the standard of practice when they receive so few specimens from patients suspected to have TB.

Under the program, ten rural PHLs perform smear microscopy onsite and simultaneously inoculate a broth culture with a portion of the specimen, which is sent to the state PHL’s Microbial Diseases Laboratory for incubation. Positive cultures undergo rapid drug susceptibility testing, including molecular testing. If physicians have reason to suspect drug resistance when the specimen is collected, the local PHL also sends sputum sediment for rapid molecular testing using a “home-brew” molecular beacon assay (developed in-house) or DNA sequencing.

This model provides for rapid availability of smear results locally, as well as access to state-of-the-art methods.

“Complacency Is Unacceptable”

With molecular assays, innovative testing models for low-incidence jurisdictions and other advances—including some in the pipeline—Desmond said he believes TB elimination is an achievable goal: “It’s not rocket science. It’s a matter of having the resources.”

But, he quickly added, “Do I believe we’re going to have the resources? I have my hopes, but maybe I’ve been around too long; I’m somewhat skeptical.”

Historically, at least, the resources have not always been available. CDC’s Castro cited “very sad evidence” of past TB funding cuts.

Between 1985 and 1992, he said, the US experienced an “unprecedented” resurgence of TB following a perfect public health storm: rising rates of HIV infection, leaving those co infected with TB especially prone to disease progression; emergence of MDR TB at a time when drug resistance testing was uncommon; lax infection control practices in institutional settings; and, importantly, a deteriorated infrastructure for TB services after the elimination of categorical federal TB funding in 1972 and subsequent state funding cuts.

“We were very ill-prepared to respond,” said Castro. By 1992, TB case rates had risen 20%.

This history perhaps explains Castro’s interest in ‘Brown’s Law’ and the aftereffects of the DIRT in 2009. “We’ve learned painfully in our past that the moment you let your guard down, it comes back to haunt you,” he said. “Complacency is unacceptable with TB.”

National TB statistics for 2010 are not yet available, but Desmond said, “I will predict for you rather than an 11% drop, we’re going to go back to the [2% to 4%] figures we were seeing before [2009], and we’re not going to be too pleased with that. This is not the point at which we should be saying TB is going to go away and we don’t need to worry about it.”

International Infection Rates Reinforce Need for TB Control Programs

Overall, 11,540 TB cases were reported in the US in 2009, but there were 9.4 million new cases worldwide. The WHO estimates that one-third of the global population is infected with M. tuberculosis.

The health officials interviewed for this article agree that TB elimination in the US cannot occur in isolation from the rest of the world, especially since the highest pockets of domestic TB are in foreign-born populations. Thus, both CDC and APHL have adopted a global perspective of TB control.

APHL is a member of the Partners’ Committee of the Global Laboratory Initiative, an independent, expert advisory group to the WHO Stop TB Program. In addition, the association:

• Coordinated the efforts of an expert workgroup to develop an exhaust cabinet that provides a safe working environment for the thousands of technicians in developing countries who prepare sputum specimens for smear microscopy—the most widely used means of diagnosing TB in developing countries.
• Implemented advanced laboratory information systems in the national TB laboratories of Kenya, Botswana and Mozambique, with more laboratories to be added. The systems connect to automated testing equipment, making it possible to test more specimens and to report results more quickly to TB clinics using text messaging services.

CDC’s Iademarco said, “I am impressed with APHL’s new strategic plan because it defines our collective mission as shaping national and global health objectives and promoting policies, programs and technologies to assure continuous improvement in the quality of laboratory practice and health outcomes. We stand behind this mission, and I think it is essential to eliminate tuberculosis.”

Despite unpredictable challenges ahead, one thing is certain in the realm of TB control. Said Castro, “TB is an infectious disease only diagnosed by finding the organism in those who have it. And therefore, the lab is at the heart of this. The ability to diagnose TB depends on having good labs.”