A big problem in preventing the spread of HIV has been that newly infected people enter the most infectious phase of the disease before HIV antibodies appear in their blood. As companies come out with new, 4th-generation tests that can spot the virus earlier, CDC has funded studies across the country to determine both the effectiveness of the new assays, as well as the best way to confirm HIV diagnoses with supplemental tests. This issue of Strategies explores a recent CDC report on these studies.

A new Morbidity and Mortality Weekly Report article on Centers for Disease Control and Prevention (CDC)-funded studies shows that 4th-generation HIV tests pick up a significant number of acute HIV infections when implemented in diverse areas of the country, the first large-scale study across populations that examines both the new assay and CDC’s new HIV diagnostic algorithm (MMWR 2013;62:489-94). The proposed new algorithm begins with a 4th-generation immunoassay that detects p24 antigen, an HIV-related protein that shows up in acute infections prior to seroconversion. Supplemental tests in the algorithm include an HIV-1/HIV-2 differentiation assay and an HIV RNA assay as a final arbiter when necessary, while Western blot and indirect immunofluorescence assays (IFA) are retired.

The MMWR article reported on two major studies of HIV testing, one in a Phoenix emergency department (ED), and another multi-site study that’s still ongoing with sites in New York City, North Carolina, and San Francisco. The researchers found that acute HIV infections detected by 4th-generation tests were often misclassified as negative by Western blot or IFA. The new algorithm averted missed diagnoses in 32% of HIV-positive patients in the Phoenix ED study and 9% of HIV-positive patients in the three-state study referred to as the STOP study, Screening Targeted Populations to Interrupt On-going Chains of HIV Transmission with Enhanced Partner Notification.

The MMWR authors noted that they were particularly surprised by the high number of acute HIV infections in the Phoenix ED study, which took place in the Maricopa Integrated Health System. The study evaluated the new algorithm through an HIV screening program that attempted to screen all adult ED patients age 18–64 for HIV who had blood work for other reasons as a part of their medical care. The overall goal of the program was to screen as many patients for HIV as possible. There were 37 undiagnosed HIV infections identified during the study, of which 12 were acute.

“While the percentage of acute HIV infections was unexpectedly high, our analysis did not specifically examine the reasons why. It is possible that HIV tests might have been ordered on some patients because of clinical suspicion, potentially increasing the number of identified infections,” said Philip Peters, MD, an author of the study with the CDC. “It is also important to note that the high percentage of acute infections evaluated through this testing program are consistent with observations that many people with acute infection develop symptoms prompting them to seek medical care. These findings also suggest that these people with nonspecific symptoms that seek care in EDs and other urgent-care settings may go undiagnosed without the use of newer testing technologies such as fourth-generation HIV tests.” Peters is a medical officer and the activity leader for HIV testing in the CDC Division of HIV/AIDS Prevention’s Epidemiology Branch in Atlanta.

The studies also solidify RNA testing as the most sensitive compared to immunoassays. For example, in the Phoenix ED study, acute HIV infections were established by a negative supplemental test but a detectable HIV-1 RNA. The median HIV-1 viral load among acute infections was 3,636,176 copies/mL, while for patients with established infections, the median was 27,125 copies/mL.

Previous research has established that 4th-generation assays can do better at detecting acute infections compared to older, antibody-only tests. However, the scale of this study dramatically outlines the human costs of less sensitive tests. “The results of these studies reported in the MMWR really show that in a program designed to identify people with acute HIV, you can actually find a substantial number of infections,” said Nicholas Moss, MD, an investigator in the San Francisco branch of the STOP study. “The studies demonstrate that there is a role for acute HIV screening in community-based testing programs serving certain populations, something that has not been shown on such a large scale before.” Moss is director of the HIV/STD section in the Division of Communicable Disease Control & Prevention in the Alameda County Public Health Department in California.

The MMWR report is also a coup for the new CDC testing algorithm, which has been praised in theory but rarely tried in practice. “I think these studies convincingly show that the HIV testing algorithm can be applied in a real-world setting and used to detect both acute and non-acute HIV infections,” Moss said. “Up to this point, although a few labs and public health departments have tried using versions of the algorithm, no one has presented a large-scale application of it. In addition, the STOP study took place in STD clinics and community-based testing organizations, and the data from Arizona is hospital-based, reflecting the utility of the algorithm in different testing settings.”

At the same time, CDC is not rushing to recommend the new algorithm be widely implemented, and remains committed to thoroughly vetting the design as study results continue to come in, according to Peters. “While this particular testing process holds promise for increasing accurate diagnosis and identifying acute infections, CDC is continuing to evaluate that process for effectiveness and has not yet released official recommendations,” he said.

One study that is ongoing is STOP, and Moss and Peters agreed there is much more to learn from the research. “Regarding diagnosing acute HIV infection, the study is comparing the performance of the fourth-generation immunoassay with pooled nucleic acid amplification testing (NAAT). Preliminary data was presented at the 2013 Conference on Retroviruses and Opportunistic Infections, and activities to evaluate the benefit of enhanced partner notification are still ongoing,” Peters said. “The ultimate goal of this project is to demonstrate a cost-effective model for acute HIV infection screening that can be adopted as a national public health program to reduce HIV transmission.”

The STOP study will also offer more data on the performance of the Multispot HIV-1/HIV-2 differentiation test, used in the new CDC algorithm as a supplemental test after a positive 4th-generation immunoassay result, Moss said. In addition, pooled RNA testing won’t be an option for many public health laboratories, so gathering more data on the comparison between 4th-generation immunoassays and RNA is vital. “In comparison to the fourth-generation immunoassays, pooled RNA is a little bit more sensitive,” he said. “The fourth-generation tests for early HIV infection may not turn positive until the viral load is around 30,000–50,000 copies per milliliter, so we’re interested to see if there are other  ways to increase the sensitivity of a fourth-generation based algorithm.”