| Sign In to gain access to subscriptions and/or personal tools. |
© 2006 SAGE Publications (B3) Markers of Immunodeficiency and Mechanisms of HAART Therapy on Oral LesionsPresented at the Fifth World Workshop on Oral Health and Disease in AIDS, Phuket, Thailand, July 6–9, 2004, sponsored by Prince of Songkla University, Thailand, the International Association for Dental Research, the World Health Organization, the NIDCR/National Institutes of Health, USA, and the University of California-San Francisco Oral AIDS Center.
1 Department of Oral and Maxillofacial Surgery, Oral Medicine and Oral Pathology, Dublin Dental School and Hospital, Trinity College Dublin, Lincoln Place, Dublin 2, Ireland Correspondence: * corresponding author, stephen.flint{at}dental.tcd.ie
Highly active anti-retroviral therapy (HAART) has revolutionized the treatment and prognosis of HIV disease and AIDS in those who can take advantage of the treatment. There are currently 20 different anti-retroviral drugs in 4 different classes that are used in specific combinations. Suppression of HIV replication and immune reconstitution are goals of therapy. Since the prevalence of some easily detectable oral manifestations of HIV/AIDS (OMHIV/AIDS) decreases with HAART, it has been suggested that they might be clinically useful surrogate markers of HAART efficacy and immune status. This might be particularly useful if their recurrence presaged or accompanied HAART failure. To date, there has been little work in this area, but its potential value to the clinical management of HIV/AIDS is apparent, especially if frequent measures of viral load and CD4 cell counts are not readily available. However, the usefulness of OMHIV/AIDS as signals for HAART failure is complicated by three phenomena: the immune reconstitution syndrome, the similarity of some adverse reactions of HAART to OMHIV/AIDS, and the direct inhibitory effect of HAART medications on some OMHIV/AIDS (e.g., inhibition of oral candidosis by protease inhibitors). This workshop considered the current evidence and proposed pertinent research questions.
Key Words: HAART • HIV • oral lesions • mucosal lesions • oral manifestations
Highly active anti-retroviral therapy (HAART) has, without doubt, changed the lives and the course of the disease in those infected with the human immunodeficiency virus (HIV). Since the "Hit Early, Hit Hard" triple-therapy concept was first introduced in 1995, there has been a striking reduction in AIDS deaths in those HIV-infected individuals who can avail themselves of treatment (Ho, 1995). The US Department of Health and Human Services and the Henry J. Kaiser Family Foundation publishes regularly updated guidelines on the use of antiviral agents in HIV-1-infected adults and adolescents. These are available at http://AIDSinfo.nih.gov. The aims of HAART are four-fold:
Successful treatment causes a sustained suppression of the virus, resulting in an undetectable viral load (< 50 copies/mL) at 4–6 months after initiation of therapy. A concomitant rise in the CD4+ count is also seen. However, HAART regimens are complex, may cause serious (potentially lethal) side-effects or drug interactions, pose difficulty with compliance, often entail large numbers of pills and tablets with frequent dosing, and carry serious potential consequences from the development of viral resistance (United States Department of Health and Human Services and Henry J. Kaiser Family Foundation, 2005). Treatment failure (HAART failure) does occur and has been ascribed to the development of resistance by the virus to the combination of drugs used, non-compliance with the drug regimen, or suboptimal potency or blood levels of the drug combination. This workshop addressed the morbidity aspect of therapeutic aim (4), since: oral co-morbidities in HIV infection were some of the first manifestations of the disease to be documented; these co-morbid oral conditions are numerous, prominent, easily diagnosed, occur early in the course of the disease, and may be virtually pathognomonic (such as oral hairy leukoplakia); and compatible classifications of these diseases have been published (Greenspan et al., 1992; EC-Clearinghouse, 1993). However, these classifications may not be entirely appropriate in resource-poor settings. Oral manifestations of HIV infection might therefore serve as good markers for monitoring not only restoration of immune function (along with the numerical parameters of viral load and CD4+ counts), but also HAART failure. These surrogate markers may give insight into the functionality of the immune system, as opposed to simple numerical biochemical and hematological counts on peripheral blood. Indeed, it has been shown that peripheral blood contains only 2% of the body’s lymphocytes, and that these are mostly naïve T-cells and NK cells, whereas the lymphoid tissues contain 98% of the lymphocytes, including activated T-cells, memory cells, and B-cells (Pakker et al., 1998; Roederer, 1998).
At the time this paper was written, anti-retroviral agents constituted 20 different drugs of 4 separate classes: fusion inhibitor (FI), nucleoside reverse transcriptase inhibitor (NRTI), non-nucleoside reverse transcriptase inhibitor (NNRTI), and protease inhibitor (PI) (Table 1
Separate guidelines have been drawn up for infants, children, and adolescents, due to unique differences in acquisition, previous antiviral exposure, diagnostic testing, age-specific immunological markers, pharmacokinetics, clinical manifestations, and considerations of adherence to drug regimes. However, principles of treatment remain the same (Working group on anti-retroviral therapy and medical management of HIV-infected children, 2005).
Question 1: What are the incidence, prevalence, and characteristics of HIV-related oral diseases in patients on, as compared with those not on, successful highly active anti-retroviral therapy (HAART)? Question 2: What is the time-course from the start of HAART therapy to any observed changes in incidence, prevalence, or characteristics of HIV-related oral diseases? Question 3: What are the incidence, prevalence, and characteristics of HIV-related oral diseases in patients who experience HAART failure? Question 4: Which oral lesions are useful as markers of immunodeficiency or HIV disease progression in patients on HAART therapy? Question 5: Which changes in incidence, prevalence, and characteristics of HIV-related oral diseases are related to immune reconstitution, and which are related to direct effects of anti-retroviral medication?
Question 1: What are the incidence, prevalence, and characteristics of HIV-related oral diseases in patients on, as compared with those not on, successful highly active anti-retroviral therapy (HAART)? Studies around the world showed various degrees of reduction in the prevalence of oral manifestations with the use of anti-retroviral therapy. In a Spanish study on 154 subjects, Ceballos-Salobrena et al.(2000) showed a 30% reduction, while Tappuni and Fleming (2001) reported a reduction of 24% in a study on 284 patients in the United Kingdom. However, a straightforward comparison between the studies is inappropriate, since the study subject groups are heterogeneous and the methodology is diverse. Some looked at a particular oral manifestation (Cauda et al., 1999) as apposed to a range of oral lesions (Ceballos-Salobrena et al., 2000; Tappuni and Fleming, 2001; Eyeson et al., 2002; Ramírez-Amador et al., 2003). Studies examining the effect of HAART on the prevalence of individual oral manifestations mainly reported on oral candidosis, oral hairy leukoplakia, HIV-related periodontal diseases, Kaposi’s sarcoma (KS), oral papillomata, and HIV-related salivary gland disease. A summary of the findings of these studies was presented in this workshop. Oral candidosis (OC) has been shown to be one of the most common oral lesions in HIV patients. With the advent of HAART, most studies reported a decrease in the prevalence of OC. In a study on 93 patients, 7% of patients on protease inhibitors (PI) had OC, compared with 36% in non-PI-treated patients (Cauda et al., 1999). Schmidt-Westhausen et al.(2000) detected OC in 10/103 (9.7%) of their study subjects who had been on HAART for 4 weeks and in none after 6 months’ therapy (n = 61) (Schmidt-Westhausen et al., 2000). Tappuni and Fleming (2001) reported that the prevalence of OC was about 50% less in patients on therapy (n = 89) compared with drug-naïve patients (n = 195). Conversely, Patton et al.(2000) found no significant difference in the prevalence of oral candidosis (OC) with the use of PI (n = 507). In the same study, the prevalence of oral hairy leukoplakia (OHL) was found to decrease with therapy (Patton et al., 2000), in agreement with reports from other studies (Tappuni and Fleming, 2001). The prevalence of HIV-associated periodontal diseases was reported to decrease significantly in an American cohort, from 4.8% to 1.7% with HAART (Patton et al., 2000), in concordance with findings in other studies (Ceballos-Salobrena et al., 2000; Tappuni and Fleming, 2001). Kaposi’s sarcoma is one of the oral manifestations that is strongly associated with HIV, although its prevalence is quite low in this group. Ceballos-Salobrena et al.(2000), looking at a Spanish HIV-positive population, found no cases of KS in patients on HAART (n = 154). Studies from the USA (Patton et al., 2000) and Mexico (Ramírez-Amador et al., 2003) found no significant change in the occurrence of KS with HAART. Unlike most other oral manifestations of HIV, studies from the USA and the UK described an increase in the prevalence of oral warts with HAART (Patton et al., 2000; Greenwood et al., 2002), which may reach statistical significance (Greenspan et al., 2001). Others looking at a different population (Mexican) reported similar detection rates of oral warts in HIV-positive subjects on HAART compared with those not on therapy (Ramírez-Amador et al., 2003). Other lesions that are showing a trend of rising prevalence include HIV-related salivary gland disease (Patton et al., 2000). However, this was not supported by other studies (Ramírez-Amador et al., 2003). There is limited evidence on the prevalence of oral lesions in HIV-infected children on HAART. In a study group of 40 children from Northern Thailand, aged 0-12 years, 12.5% were on HAART and 57.5% had oral lesions (Pongsiriwet et al., 2003). In 102 American vertically infected 3- to 15-year-olds on HAART, 2.9% had OC, 2.9% linear gingival erythema, 2% bilateral parotid enlargement, and 1% enlarged cervical lymph nodes (Okunseri et al., 2003). No OHL, oral papillomata, or KS was detected. The effect of therapy on oral manifestations in 38 HIV-positive American children was studied by Flanagan et al.(2000), and Khongkunthian et al.(2001) compared the prevalence of HIV-related oral lesions between Northern Thai children on HAART (n = 15) and children not on HAART (n = 30). Both studies reported no significant difference in HIV-related oral lesions in children on HAART compared with children not on therapy. Published studies agree that the overall prevalence of HIV-related oral lesions decreases with therapy. However, some lesions, such as oral warts and salivary gland disease, are showing a trend of increasing with HAART. Further longitudinal studies are needed to confirm these findings.
Question 2: What is the time-course from the start of HAART therapy to any observed changes in incidence, prevalence, or characteristics of HIV-related oral diseases?
Incident oral opportunistic infections have been reported to occur only in patients who receive HAART if there is regimen adherence failure, or non-compliance (Manfredi, 2004). However, HAART itself has been associated with adverse reactions (Scully and Diz Dios, 2001), which may occur at any time after initiation of therapy, and some of these conditions are also seen in untreated HIV disease. It is therefore important to distinguish between a true HIV-related oral manifestation of disease and adverse events of HAART. For example, the NRTIs have been associated with intra-oral ulcers in 3–30% of patients receiving zalcitabine (ddC) (McNeely et al., 1989), with mucocutaneous hyperpigmentation in patients receiving zidovudine (AZT) (Greenberg and Berger, 1990; Poizot-Martin et al., 1991), and with xerostomia in 30% of patients taking didanosine (ddI) (Allan et al., 1993). These adverse effects must be distinguished from "Not Otherwise Specified" ulcers (NOS), aphthous stomatitis, mucosal pigmentation, and xerostomia associated with HIV disease in the absence of HAART (Greenspan et al., 1992; EC-Clearinghouse, 1993). Xerostomia and/or oral ulcerations can also be seen in 7% of patients taking protease inhibitors (Danner et al., 1995). Other adverse events include taste abnormalities in 20% of patients (Schiffman et al., 1999), circumoral paraesthesia in a quarter of patients receiving ritonavir (Scully and Diz Dios, 2001), parotid lipomatosis or facial lipoatrophy (as a feature of generalized lipodystrophy) in patients receiving ritonavir (Olive et al., 1998), and cheilitis in 57% of patients taking indinavir (Calista and Boschini, 2000). Although NNRTIs have rarely been implicated in HIV-associated disorders, erythema multiforme with nevirapine treatment has been reported (Wetterwald et al., 1999; Fagot et al., 2001). As to oral manifestations in long-term HAART and older patients, studies on an association with osteoporosis or osteodystrophy are inconsistent. Madeddu et al.(2004) described an increase in osteoporosis in patients with HAART, whereas Thomas and Doherty (2003) reported that osteoporosis may occur in patients with and without HAART. If this association is real, there may be implications for alveolar bone loss or periodontal disease (Yoshihara et al., 2004). Similarly, HAART-induced diabetes mellitus and dyslipidemia may predispose to periodontal disease in the long term (Cutler and Iacopino, 2003; Renvert, 2003). A disproportionate increase in the incidence and prevalence of neoplasms like lymphoma and carcinoma (Manfredi et al., 2003a; Manfredi, 2004), due to persistent dysregulation of the immune system, has been surmised. One consistent observation, however, is an increase in benign HPV-associated oral neoplastic lesions (papillomas, condylomas, and focal epithelial hyperplasia) in patients on HAART (Patton et al., 2000; Schmidt-Westhausen et al., 2000; Greenspan et al., 2001). This raises the possibility of an increased incidence of HPV-related oral malignancies with long-term HAART, particularly if the HPV subtypes carried are high-risk subtypes (such as HPV-16) which have been associated with oral malignancy.
Question 3: What are the incidence, prevalence, and characteristics of HIV-related oral diseases in patients who experience HAART failure? Virological failure manifests as failure to achieve HIV RNA levels of < 400 copies/mm3 by 24 weeks, or < 50 copies/mm3 by 48 weeks (although ideally by 24 weeks) in a treatment-naïve patient, and suggests an incomplete response to HAART. However, account must be taken of the fact that high baseline levels may affect the time-course and the slope of the monitored virological response. Most patients with an adequate virological response have at least 1 x log10 copies/mm3 reduction in viral load at 4 weeks. Immunological failure represents a failure to increase the CD4+ count by 25–50 cells/mm3 per month above baseline in the first year, or a decrease in CD4+ count below baseline levels. Low baseline CD4+ count may be associated with a slower response to therapy. Successful therapy achieves a mean increase of 150 cells/mm3 in the first year. However, absolute cell counts take no account of the functionality of the CD4+ cells. Clinical failure is defined as the occurrence of HIV-related clinical events after 3 months of HAART therapy, and thus, in combination with absolute cell counts, gives an indication of the functionality of the cells measured. Since many oral manifestations of HIV occur early in the course of HIV disease, these may become important markers of early HAART failure. Several publications (Pakker et al., 1997; Gorochov et al., 1998; Notermans et al., 1999) have indicated a return of immune function with HAART, but there is less information about the different combinations of HAART and specific functionality against oral pathogens. A single report (Nielsen et al., 1994) assessed cell-mediated lymphocytic function against Candida using mitogen stimulation. Skolasky et al.(2001) have suggested that parameters such as presence or absence of fever and thrush (including oral pseudomembranous candidosis) may represent markers of immune competence with HAART. A further complication in defining clinical failure is the so-called immune reconstitution syndrome—or ‘HAART Attack’ (Kunimoto et al., 1999). The re-invigorated inflammatory component of immune reconstitution can have a deleterious effect on quiescent or latent infection, whereby recrudescence of the disease causes damage to the host. The immune reconstitution syndrome (French, 1999) has been described for infections with: tuberculosis, cerebral cryptococcosis, cytomegaloviral retinitis or uveitis, Pneumocystis carinii pneumonia, microsporidial keratoconjunctivitis, and visceral leishmaniasis. The oral equivalent is perhaps represented by the reported increase in HPV lesions (Greenspan et al., 2001) and salivary disease, particularly the diffuse infiltrative lymphocytosis syndrome (DILS) (Patel and Mandel, 2001). It is obviously important that immune reconstitution syndrome be excluded before clinical failure is diagnosed. There is a growing body of evidence that HAART reduces oral manifestations of HIV infection (Margiotta et al., 1999; Ceballos-Salobrena et al., 2000; Patton et al., 2000; Schmidt-Westhausen et al., 2000; Tappuni and Fleming, 2001; Eyeson et al., 2002; Greenwood et al., 2002; Greenspan et al., 2004; Kerdpon et al., 2004; Nicolatou-Galitis et al., 2004), but none of these investigations addressed the issue of recrudescence of oral manifestations of HIV in HAART failure. There is a single case report of multiple opportunistic AIDS-associated disorders that paralleled immunodeficiency levels in a girl with congenital HIV infection, which documented opportunistic infections following virological failure (Manfredi et al., 2003b). Since many oral manifestations of HIV occur early in the disease, these may become important markers of early HAART failure in the future, but at the time this paper was written, there were no cohort studies in the literature. It is vital, however, that any future studies take account of the definition of HAART failure and thereby have the same inclusion or end-point criteria.
Question 4: Which, if any, oral lesions are useful as markers of immunodeficiency or HIV disease progression in patients on HAART therapy?
Although many oral lesions were recorded in HIV disease prior to the advent of HAART therapy, not all of those described decreased upon the advent of HAART. The prevalence of several oral lesions (oropharyngeal candidosis [OPC], oral hairy leukoplakia [OHL], Kaposi’s sarcoma [KS], and HIV-associated periodontal diseases) decreased in the presence of HAART, suggesting that these lesions are the most sensitive to immune reconstitution and/or decrease in HIV viral load (Patton et al., 2000; Schmidt-Westhausen et al., 2000; Greenspan et al., 2001; Tappuni and Fleming, 2001). Of these four lesions, the most two most common, both pre- and post-HAART, are OPC and OHL (Patton et al., 2000; Schmidt-Westhausen et al., 2000; Greenspan et al., 2001, 2004; Tappuni and Fleming, 2001). Both candidosis and OHL can occur early in the course of HIV disease, with CD4+ T-lymphocyte numbers as high as 400 cells/mm3, which has made them good indicators of early immune dysfunction. However, unlike OPC, OHL has a greater prevalence in males than females (Shiboski et al., 1996; Patton et al., 2000). Candidosis has the added diagnostic benefit that it can present in one of two clinical forms, depending on CD4+ cell count. Erythematous candidosis presents at higher CD4+ cell counts (~ 400 cells/mm3) compared with the pseudomembranous form, which more commonly occurs when CD4+ cells decrease further (~ 200 cells/mm3) (Greenspan et al., 1992). Studies have also demonstrated a significant correlation of OPC and increased viral load (Campo et al., 2002; Greenspan et al., 2004). However, an increase in erythematous candidosis was observed in a Spanish cohort, despite a decrease in the pseudomembranous form (Ceballos-Salobrena et al., 2000). Thus, OPC fulfils three of the criteria for a useful marker of HIV disease progression following HAART failure, in that it is not gender- or race-specific, it occurs early in immune dysfunction in the erythematous form, and the prevalence correlates to HIV viral load. Confounding this, however, is the impact of protease inhibitors on both the carriage of Candida albicans and the occurrence of OPC (Cassone et al., 2002), suggesting that there may not be an increased incidence of OPC in individuals receiving PIs, even in a failing regimen.
Question 5: Which changes in incidence, prevalence, and characteristics of HIV-related oral diseases are related to immune reconstitution, and which are related to direct effects of anti-retroviral medication? This apparent ability of protease inhibitors to inhibit growth of Candida may be related to similarities between candidal-secreted aspartic proteinases (SAPs) and HIV proteinase (Munro and Hube, 2002). SAPs are key virulence factors for Candida albicans, and their highly conserved active-site sequences (Asp-Thr/Ser-Gly) and 3-D structures closely resemble those of HIV proteinase (also an aspartic proteinase, with the same active-site sequence). In vitro studies found that therapeutically relevant concentrations of indinavir, saquinavir, and ritonavir could inhibit some of the SAPs involved in Candida virulence, suggesting that a fortuitous cross-reaction may enable HIV proteinase inhibitors to prevent episodes of clinical candidosis directly. Further evidence comes from prospective and cross-sectional clinical studies (Cassone et al., 2002). In the prospective study, two groups of HAART-naïve patients began either a PI-based HAART or a NNRTI-based HAART regimen and were followed for 180 days. Significant decreases in episodes of oral candidosis and Candida carriage were observed in the PI group but not in the NNRTI group. Additionally, there was a rapid decrease in SAPs isolated from saliva of Candida carriers in the HAART-PI group but not in the HAART-NNRTI group. That there was no significant difference in CD4+ cell count increases or HIV viremia decreases between the two groups during study suggests that the decreases in oral candidosis and Candida carriage were due to a direct effect of PI use, rather than to differences in the extent of immune system reconstitution. In the cross-sectional study, Candida carriers who had been on a stable HAART-NNRTI or HAART-PI regimen for a year were examined. Those on HAART-NNRTI had SAPs in their saliva, but those on HAART-PI did not, although there was no significant difference between regimens as to CD4+ cell count increases and HIV-viral load decreases. This suggests that the effect of PI use on Candida virulence is independent of immune reconstitution. Whereas a decrease in episodes of candidosis benefits HIV-seropositive patients, PI use may lessen the usefulness of oral candidosis as a marker for HAART success or failure.
In opening the workshop to participation by those in attendance, a great many issues were raised. The research questions which were generated by the discussion are summarised in Table 2  .
It was generally agreed that the area examined by this workshop would be a fruitful field for future research. The four main themes in Table 2  were considered by the workshop to be those questions on which the profession might focus future research effort, since these were important questions which may have clinical relevance and significant influence on patient care. Several participants stressed that, in future observational studies (in both resource-rich and resource-poor settings), if we are to monitor oral manifestations in HIV in HAART-treated patients, it is critical that the definition of HAART failure be consistent (and if not consistent, documented). Furthermore, criteria for entry into and cut-off in studies should be carefully documented and raw data retained, in case the need arises to re-work the data, should criteria or definitions change in the future. In the collection of tracking data, documentation of not just current drug use but also past drug history is important, because the analysis will be very complex, since there are 4 different drug classes and 20 different drugs. Factors that may affect current observations are previous drug therapy, compliance, drug holidays, and toxicity. Finally, increased collaboration with medical colleagues was seen as highly desirable, and all efforts should be made to have a dental input to future major studies of HAART.
Advances in Dental Research, Vol. 19, No. 1,
146-151 (2006) This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
Abstract
Introduction
