Treating PML: hopes for future focus on mastering immune response

Understanding how the JC virus and immune responses to it cause the devastating central nervous system disorder PML (progressive multifocal leukoencephalopathy (PML) and the related PML immune reconstitution inflammatory system (PML-IRIS) could lead to better ways of managing the conditions.

At last month’s Second HIV Infection and Central Nervous System: Developed and Resource-Limited Settings meeting in Venice, neurologists and researchers heard that specific aspects of the immune response to JC virus protect against PML and lead to better survival, but that other very closely related immune responses could aggravate the condition, leading to PML-IRIS. Data suggest that both forms of PML may be equally life-threatening. (See related article on PML IRIS)

Determining which specific parts of the immune response to the JC virus to switch on and switch off may hold the best hope of controlling this horrible and hard-to-treat opportunistic infection in HIV-positive people.

JCV is a very common virus which has infected almost everyone on the planet — but it very rarely causes any symptoms or complications — even in people who are immune-compromised. However, in an unfortunate few with suppressed immune systems, this virus can take hold and begin replicating within the central nervous system, causing terrible damage known as PML. Researchers have long tried to understand why some people develop this syndrome and others don’t.

What seems clear from the data is that high levels of JCV replication in the brain can occur when the immune surveillance has weakened, and that, without treatment, higher levels of JCV DNA in the CSF are associated with worse PML outcomes. However, the best treatment for PML developed thus far, at least for people with HIV-associated PML, is highly active antiretroviral therapy (HAART). (See related article on the incidence, diagnosis and treatment of PML in the HAART era).

Good CD8 cells

In addition, some of the CD8 immune response may be beneficial in PML. “There is also a good side of the immune response against JC virus,” said Dr Koralnik. Data from several studies, including from Dr Annamaria Pazzi, of the San Raffaele Scientific Institute in Milan, have demonstrated that lower JCV-CSF levels were associated with better outcomes — and most data suggest that the cytotoxic T lymphocyte (CTL) response may be necessary to suppress JCV in the brain.

In fact, Dr Koralnik shared data from his own lab showing that there are JCV-specific CD8 cells CTLs that target certain epitopes (amino-acid sequences) of JCV’s viral protein 1 (VP1). Looking at stored samples taken from the AIDS bank repository and other cohorts, Dr Koralnik discovered that these VP1-specific CTLs can be found in 91% of PML survivors versus 9% of PML progressors (in 1/15,000 -1/24,000 PBMC). This was a marker of favourable outcome that was detectable in the CSF of PML survivors and also at low levels in healthy individuals (in 1/100,000 PBMC) but rarely in PML progressors.

Because of such findings, Dr Koralnik believes there could “be some hope for the setting of immune therapy for PML”, such as adoptive transfer therapy, dendritic cell-based immunotherapy, designer T-cell therapy or perhaps vaccine strategies.

Bad CD8 cells

Ironically, Dr Avindra Nath of Johns Hopkins University, Baltimore and others presented data at the conference showing that when the immune response is suddenly restored by HAART, cytotoxic T lymphocyte (CTL) CD8 cells that return to the brain to find evidence of a JCV infection (which may not have caused any clear symptoms), often go into overdrive, killing virtually any cell with evidence of JCV, and perhaps many innocent bystander cells as well. For all intents and purposes, the PML-IRIS condition that this causes is clinically indistinguishable from classical PML. (See related article on PML IRIS).

After hearing this, some of the other scientists in the audience asked Dr Avindra Nath of Johns Hopkins University, Baltimore, whether immune modulating treatments that would down-regulate the cytotoxic CD8 response would help or hurt people with IRIS.

“I think it is a little too early to know what that regimen would be,” he responded. “But I think it is exciting to think that you have predominantly CD8 cells there. In our patients that I presented, we used steroids, now steroids are probably not the most ideal choice of drugs. But now that we have so many new immunomodulatory drugs we can regulate only effector T-cells or subsets of T-cells, so I think that soon we may have the ability to modulate the immune system whatever way we want.”

However, Dr Igor Koralnik of Beth Israel Deaconess Medical Center and Harvard Medical School, one of the world’s leading PML experts, described a cautionary tale of what can happen when you tinker with the immune system a little too much.

Natalizumab (Tysabri), a humanised monoclonal antibody against alpha 4 integrins, is an extremely promising new medication in clinical development for patients with multiple sclerosis (MS) and Crohn’s disease. It is given once a month intravenously with biological activity lasting up to three months. The activity of this compound is very specific — it keeps lymphocytes and monocytes in the blood stream and prevents them from trafficking to the brain, gut and other organs.

Unfortunately, the drug had to be withdrawn voluntarily by its maker, Biogen Idec, in February 2005 after two cases of PML were discovered in MS patients. A retrospective analysis of natalizumab-treated patients found that one case of PML had also developed in a patient with Crohn’s disease. Several studies were done, and the drug was reinstated in the summer 2006 after the risk of PML was determined to be about one out of 1000 patients over 18 months of follow-up.

According to Dr Koralnik, one of the patients had “one of the most aggressive forms of PML anyone has ever seen." Histopathology studies could find no inflammatory infiltrates within the PML lesion, he said “but there were billions of JC virus-infected cells in the brain. So, natalizumab did a good job of preventing bad lymphocytes that wanted to cause MS from going into the brain of this patient but it did an even better job at preventing good lymphocytes that wanted to prevent reactivation of latent viruses [from getting there], which led to this catastrophic PML presentation.”

In another one of the patients, PML was caught early and natalizumab was discontinued. However, three months after stopping natalizumab, “the patient developed a very inflammatory reaction, just the same as PML IRIS seen in HAART… that nearly killed the patient,” said Dr Koralnik. “And I think that natalizumab is only the tip of the iceberg. Every biopharmaceutical company known to man has in the pipeline new medications that have the same philosophy as natalizumab, to sequester some lymphocytes in the bloodstream, and in the lymph nodes in the blood and prevent them from going into the tissue and doing damage. This is something of course we need to follow, especially in a resource-developed setting.”

The virus

The particular target of the CD8 cells could be telling. It’s worth noting that JCV uses VP1 to enter glial cells in the CNS. One study, by Dr Serena Delbue, was conducted to determine whether some JCV variants were more likely to be associated with more aggressive PML in the brain.

But what her group found was just the opposite: people diagnosed with PML with slower courses of clinical progression had actually developed JCV variants with certain amino acid changes in the VP1 loop that were associated with lower CSF levels of JCV. It is likely that JCV with these polymorphisms are escape mutants that evolved to evade the CTL CD8 cells — but which in turn make them less fit.

In contrast, Dr Koralnik reported on a case study of a person with HIV and JCV infection which led to neuron loss and cerebellar atrophy. Subsequent examination of the patient’s JCV found a different change in the VP1 that alters the viral tropism, allowing JCV to actually infect granule cell neurons.

However, studies of the JCV VP1 could also lead to treatment strategies — what would seem to be most important is to find ways of keeping the virus out of the brain. So, one recent intriguing discovery is that JCV’s VP1 uses the 5-HT2a> receptor — a serotonin receptor — to enter glial cells.

This raises the possibility that serotonin blockers might have a role in treatment for PML. According to Dr Koralnik, case reports will be coming out showing favourable outcomes with one, mirtazapine. His own site has tried the same approach with mixed results, although he concedes that they may be using what amount to “homeopathic” doses. Some other serotonin blockers that are not available in the US (such as ketanserin, which has been shown to cause hypotension) and ritanserin (which increases the QT interval) may be more potent inhibitors.

Used to disappointment, though, Dr Koralnik believes these drugs might, at best, be useful as an adjunct therapy. “We can hope at best for containment of the virus, but not the cure. But since there is no other good option, it can’t hurt our patients and most times they need an antidepressant medication. There is no contra-indication to giving them this medication while they are on HAART and so this is something to follow. Maybe there will be more potent blockers in the future that may help in the containment of PML.”

See related article on the incidence and diagnosis of PML in the HAART era.