Non-Hodgkin's lymphoma risk increased with prolonged high HIV viral load

Prolonged high HIV viral loads are associated with an increased risk of non-Hodgkin's lymphoma (NHL) independent of CD4 cell counts, according to data from the French Aquitaine Cohort published in the February 1^st edition of Clinical Infectious Diseases. The investigators found that the risk of NHL decreased after six months or longer of highly active antiretroviral therapy (HAART). In addition, their data suggest that a history of hepatitis C infection may be protective against NHL, and they conclude that this warrants further investigation.

Non-Hodgkin's lymphoma is one of the most common causes of HIV-related mortality in industrialised nations, and in France it accounts for between 11%-15% of all deaths in HIV-positive individuals and for 23%-30% of AIDS-related causes of death. Previous studies have found that in comparison with the general population the risk for NHL is 100 to 400 times higher in people infected with HIV.

In order to ascertain whether there is a link between NHL and infection with various herpes viruses, as well as hepatitis C virus (HCV), investigators from France performed a case-control study using the Aquitaine Cohort database (including Bordeaux University Hospital and four other public hospitals in Aquitaine, southwestern France). There were 52 identified cases and 145 controls, individually matched for gender, CD4 cell count, calendar period of enrolment, and follow-up duration.

Sixty-five new cases of systemic B cell NHL were actually identified between 1996 and 2003, although only 52 fulfilled the study criteria by being diagnosed at least three months after joining the cohort. Lymphoma subtypes, according to WHO criteria, included diffuse large B cell lymphoma (46%), Burkitt-like lymphoma (13%), and anaplasic (4%). Another 13% were untyped and 24% were unknown.

Increasing incidence

The investigators found that incidence of NHL increased during the study period. Between 1996-1999 there were 3.05 cases per 1000 patient-years (95% CI, 1.96-4.14 cases per 1000 patient- years). Between 2000-2003, there were 3.35 cases per 1000 patient-years (95% CI, 2.24-4.47 cases per 1000 patient-years). The mean incidence rate of NHL throughout the study period was 3.21 cases per 1000 patient-years (95% CI, 2.43-3.99 cases per 1000 patient-years).

Does hepatitis C infection reduce NHL risk?

A total of 21% of case patients tested positive for HCV antibodies during the study period compared with 39% of the controls. The OR for NHL was 0.35 (95% CI, 0.16-0.76) for those with HCV antibodies compared to those without. However, when missing data was included (including 5 case patients and 18 controls), the OR increased to a non-significant 0.60 (95% CI, 0.20-1.83). Nevertheless, in multivariate analysis, history of HCV infection was found to be associated with a decreased risk of NHL (OR, 0.34; 95% CI, 0.14-0.78).

The lack of an association with an increased risk of NHL in HCV-coinfected patients has been suggested by several studies in non-HIV infected patients. However, this is the first study to show that HCV coinfection could be associated with a decreased risk of NHL, "although," the investigators write, "missing data did not allow us to make a formal conclusion on this association. Thus, the relationship between HCV infection and NHL warrants further study before any conclusion can be suggested."

No association with herpes viruses

History of infection with herpes simplex virus, varicella zoster virus, cytomegalovirus or of Kaposi's sarcoma was not found to be associated with AIDS-related NHL (OR, 1.49; 95% CI, 0.78-2.84). A total of 6% of case patients and 13% of controls received antiherpetic therapy for at least six months (OR for treatment vs. no treatment, 0.40; 95% CI, 0.11-1.44). The investigators conclude that they found no strong evidence to support the hypothesis that herpes viruses are associated with the development of NHL, "despite a tendency for the case patients to have received less antiherpetic drugs than the control subjects."

Increased NHL risk for those with sustained high HIV viral loads not on HAART?

"Our findings raise an important question about HAART," they conclude, "because most recent studies have been in favor of delaying HAART initiation during chronic HIV infection, considering the increasing occurrence of iatrogenic events and the absence of clear clinical, immunological, or virological advantages in starting HAART for patients with a CD4+ cell count > 350 x 10 cells/L. This long period before HAART initiation with persistent HIV replication and B cell stimulation could induce a sustained risk of NHL, justifying a close monitoring of these patients."

Immune status and HAART

The investigators did not observe a significant relationship between the CD4 or CD8 cell count nadir and the risk of NHL. However, a total of 52% of case patients and 65% of controls received HAART for at least six months. They found that six months or longer of HAART produced a significant protective effect (OR, 0.46; 95% CI, 0.21-0.98). This remained significant in multivariate analysis (OR, 0.37; 95% CI, 0.16-0.87).

"This study confirms the favorable impact of HAART on the risk of occurrence of NHL," write the investigators, "but also confirms that a minimum duration of six months of HAART is required to prevent NHL. It also suggests HIV RNA level as a possible risk factor for AIDS-related NHL in the era of HAART, independent of immunodeficiency."