P24 immunohistochemistry on lymphoid tissue: the histopathologist’s role in HIV diagnosis

In an associated original research paper published in this issue of Histopathology we have demonstrated the value of p24 immunohistochemistry on paraffin-embedded sections of surgical biopsy specimens of lymphoid tissue in the diagnosis of human immunodeficiency virus (HIV) disease. Most histopathologists in the UK are aware of the existence of this immunostain, but it is rarely used in the diagnostic setting, despite the well-recognized medical and social advantages of early diagnosis of HIV infection. Anecdotal experience suggests that this relates to confusion surrounding the anomalous position of testing for HIV infection in tissue and the question of consent. In this review we address the ethical and political issues relating to immunohistochemical testing for HIV on diagnostic biopsy specimens and advance the case for normalization of the usage of p24 in diagnostic histopathology. It is now a quarter of a century since the emergence of human immunodeficiency virus (HIV) in the UK. Since then, although histopathologists, like their colleagues in other specialties across the breadth of medicine, have exponentially expanded their knowledge, understanding and recognition of the protean manifestations of the disease, it could be argued that they have side-stepped a fundamental, but thorny issue: that of the histopathological diagnosis of HIV itself. A monoclonal antibody to HIV p24 antigen, the viral major core protein constituting part of the Gag complex, that is effective on paraffin wax-embedded tissue sections has been available since the mid 1980s,1 and yet experience from referral material to a specialist centre for infectious disease and haematopathology suggests that this immunostain does not feature in the palette of the majority of histopathology departments in the UK. In the accompanying original research paper published in this issue of Histopathology we have demonstrated the value of p24 immunohistochemistry on lymphoid tissue in the diagnosis of HIV infection, and in this paper we review the relevant current UK guidelines and argue for the expansion of its use. The manifestations of HIV infection are protean and include lymphoid hyperplasia and hypoplasia in lymph nodes and other organs, malignancies, notably lymphomas, opportunistic infections and an expanding range of degenerative organopathies. The changing morphological appearances in HIV lymph nodes as the disease progresses are now well described in all standard haematopathology textbooks. The first stage is of follicular hyperplasia, with expanded, serpiginous germinal centres sporting a well-developed follicular dendritic cell (FDC) network and numerous tingible body macrophages. The next stage is of follicle ‘lysis’, characterized by depletion of the mantle zones and small germinal centres with scarce B lymphocytes, an incomplete FDC network and infiltration by CD8 T-suppressor lymphocytes. By the end stage of HIV disease there is complete depletion of lymphoid follicles (in parallel with the very low peripheral blood CD4 count) and replacement by a stromal vascular proliferation and an infiltrate of immunoblasts and plasma cells. Monocytoid B-cell hyperplasia and multinucleated giant cells with the immunophenotype CD68 + , S100−, CD21/35− may be seen in the follicular hyperplasia and lysis stages. All these changes are well known to histopathologists and should provoke suspicion of HIV infection, even though non-specific in the early stages of infection, with a significant morphological overlap with the various other causes of follicular hyperplasia. Lymphoid changes in extranodal sites and in other organs may be more difficult to associate with HIV infection, but lymphoepithelial cysts of the salivary glands, Burkitt lymphoma, other extranodal lymphomas and Hodgkin’s lymphoma, especially when associated with Epstein–Barr virus infection, should raise suspicion. Likewise, unusual and opportunistic infections should alert the histopathologist to the possibility of underlying HIV disease. The p24 antigen is the major core protein constituting part of the HIV Gag complex, and its detection is used in routine serological testing for HIV. A monoclonal p24 antibody specific for HIV-1, which could be successfully applied to formalin-fixed paraffin wax-embedded tissue, was developed in the mid 1980s (clone H-11, marketed as clone Kal-1).1,2 In lymph node sections the positivity is localized to the FDCs within germinal centres.3–7 Also highlighted are a minor, scattered population of interfollicular paracortical lymphocytes and the multinucleate giant cells characteristically seen in nasopharyngeal/tonsillar tissue.8,9 In the brain, p24 immunohistochemistry highlights the microglial cells (single and multinucleate) of microglial nodules.10 Immunoreactivity is cytoplasmic in all cases. p24 expression has been demonstrated immunohistochemically in lymph nodes of untreated HIV-infected individuals from soon after primary infection (when clinically asymptomatic) up until the terminal stages of HIV disease, when the FDC meshwork has been destroyed. Even in patients on highly active antiretroviral therapy (HAART), p24 has been detected immunohistochemically as long as 36 months after initiation of treatment:11–13 p24 can be demonstrated by immunohistochemistry on sections of lymph node even when the blood viral load is undetectable.14–16 A small number of our cases from patients known to be HIV+ failed to show p24 immunoreactivity in their lymph nodes on immunohistochemistry, and the possible reasons for these false negatives are discussed in our accompanying research paper in this issue of Histopathology. Neither in the literature nor in our cohort have we encountered any false-positive cases. As have others, in that study we have, in addition, demonstrated p24 positivity in reactive follicular lymphoid tissue forming in non-lymphoid organs of HIV-infected patients, such as lung, anus and parotid gland. Isolated positive cells may also be demonstrated in bone marrow trephine biopsy specimens (unpublished observations). Several publications by histopathologists in the last 25 years have explored p24 expression by immunohistochemistry in paraffin sections of tissue specimens procured at surgery for diagnosis, encompassing lymph nodes,15,17–25 tonsil/Waldeyer’s ring,8,9,26–29 thymus,30 salivary gland,31–35 brain/spinal cord,36–42 retina,43,44 skeletal muscle,45 stomach,46 bowel,47–49 liver,50 placenta51–53 and others.54 However, to the best of our knowledge only two of these studies have explored the use of p24 immunohistochemistry to diagnose HIV infection. The first was published in 1989 and related to salivary gland lymphoepithelial cysts.35 Almost 20 years later Russano de Paiva et al.25, from Toulouse, France, published a landmark article in the American Journal of Surgical Pathology, entitled ‘Discovery of human immunodeficiency virus infection by immunohistochemistry on lymph node biopsies from patients with unexplained follicular hyperplasia’. By using the word ‘discovery’, their paper was the first to suggest that the histopathologist, using immunohistochemistry, can and should diagnose HIV on tissue samples. Since the first cases of HIV in the UK in the early 1980s the disease has been accorded a status unparalleled in the history of modern medicine, and we suspect that this has laid the foundation for the persisting reluctance of histopathologists, 25 years later, to maximize their potential in the diagnosis of HIV. A Department of Health national television advertising campaign in the mid 1980s depicting, for example, the eerie image of an iceberg accompanied by chilling music instilled fear in the population about the disease. Uniquely, it was mandated that specific consent be obtained from the patient before serological testing for HIV and, moreover, that the patient should receive ‘pre-test counselling’. Simply having had an HIV test (regardless of outcome) adversely affected the individual’s insurance policy status. Extraordinary measures were advocated to protect the confidentiality of a patient’s HIV status, even within the medical profession, which, it may be inferred, have led to the still prevalent practice of coyly stating the CD4 count or at best ‘retrovirus infection’ in lieu of any explicit mention of HIV, on histopathology request forms. Fortunately, times have changed, and with the introduction of HAART, there has been a two-thirds reduction in death from HIV disease since 1995 in the UK,55 despite an increase in prevalence from approximately 25 000–77 000 (UK Health Protection Agency website, http://www.hpa.org.uk). HAART can render a sustainable depletion in the viral load to undetectable levels and nurture the CD4 count back towards the normal range. What started as an often rapidly fatal disease, associated with almost invariably late presentation, has, in this country, become a chronic, but most likely eventually fatal, one, with patients now being granted a decade or more of life free of symptoms. So the 2006 UK National Guidelines on HIV testing55 referred instead to a ‘pre-test discussion’ and stated that ‘The benefits from knowing the diagnosis, and thus being able to act on this, have clearly been shown to improve the life expectancy of the individual and may outweigh many other issues’. These benefits are listed as ‘improvement in the health and well-being of individuals through access to medicines; improvement in the Public Health from the expected reduction in onward transmission; and patient empowerment in knowing their status’. In June 2007 an editorial in the British Medical Journal56 and subsequent letter57 fanned the winds of change, joining the growing medical voice calling for the end of ‘HIV exceptionalism’ and an expansion of testing. Resonating with this, the 2008 UK National Guidelines on HIV Testing have widened the net, recommending that ‘An HIV test should be considered in the following settings where diagnosed HIV prevalence in the local population (Primary Care Trust/Local Authority) exceeds two in 1000 population: all men and women registering in general practice; all general medical admissions. … HIV testing should also be routinely offered and recommended to the following patients: all patients presenting for healthcare where HIV, including primary HIV infection, enters the differential diagnosis..’.58 Frustration at the lack of implementation of these guidelines and the resulting persisting ‘extraordinary failure in public health’ in relation to HIV disease in the UK is expressed in an editorial in The Lancet published in May 2009.59 The second of the two major general medical journals of this country has thrown its hat into the ring. Turning to the role of the histopathologist in the diagnosis of HIV, the guidelines published by the UK General Medical Council (GMC) and Royal College of Pathologists (RCPath), which have some bearing on the issue, are extracted below. The GMC’s June 2008 publication, ‘Consent: patients and doctors making decisions together’,60 states the following: “Making decisions. The scope of decisions. 37. Note for pathologists and radiologists: there may be times when uncertainty about a diagnosis can only be resolved by investigations which were not specifically ordered as part of the original request for testing. If these investigations appear to fall outside the scope of the original consent given by the patient, or there are particular sensitivities around the condition for which the pathologist or radiologist wishes to test, they must contact the treating doctor and establish whether further discussion with, and consent from, the patient is necessary before proceeding. …. Expressions of consent. 49. You should also get written consent from a patient if:… (b) there may be significant consequences for the patient’s employment, or social or personal life …”. (our italics) Of the RCPath’s published guidelines, two are of particular relevance to this matter. The 2002 Good medical practice in pathology publication61 states: “21. You must not advise or recommend an investigation or treatment which you know is not in a patient’s best interests or deliberately withhold appropriate investigations or referral.” The March 2009 Guidance on consent for the processing and analysis of clinical samples following an initial consultation62 states: “2. Consent 2.1.., It is important to remember that the initial consent is likely to involve a request to investigate what is wrong, rather than to perform a specific set of analyses. … 3. Analysis 3.2 If it is believed that investigations should be performed which appear to fall outside the scope of the original consent given by the patient, or there are particular sensitivities around the condition for which the pathologist wishes to test, the General Medical Council states that the pathologist must contact the treating doctor and establish whether further discussion with, and consent from, the patient is necessary before proceeding... 4. Principles 4.1 All healthcare professionals accept that the purpose of an investigation is to solve a clinical problem, the origin of which is a consultation at which valid consent was obtained. 4.2 If consideration is to be given to extending the investigation beyond the list of tests that have been specifically discussed, then those responsible must satisfy themselves that either • the course of action lies within the overall nature of the problem or • the information revealed by the results available requires immediate further investigation because of the clinical importance of the situation and obtaining further explicit consent from the patient is impractical and • the investigation is in the best interests of the patient. 4.3 Patients have the right to exclude performance of specific tests. 4.4 Any important information obtained during an investigation cannot be ignored. It is a general principle that such information, however obtained, be explained to the patient by an appropriate member of the investigating team, which includes clinicians and pathologists. … 4.5 If it is apparent that a field of enquiry has been opened up that is remote from the original investigation, the requesting clinician should be contacted to consider the advisability of obtaining further consent. This is a matter of professional judgment and healthcare professionals must be prepared to defend any decisions taken on the patient’s behalf as being in the patient’s best interests and that the interests being protected are important enough to outweigh the normal objective of providing maximum autonomy for the patient” (our italics). The contradictions and tonal inconsistencies in the above guidelines are evident. Whilst none of the above guidelines explicitly mentions HIV disease, much less the role of HIV p24 immunohistochemistry in the diagnosis of HIV disease, the question falls within the general consideration of the histopathologist’s role in tissue diagnosis. The very essence of the bulk of histopathology diagnostic work is the diagnosis of an often fatal disease, cancer, which may have ‘significant consequences for the patient’s employment, or social or personal life’. However, none of us would dream of not rendering this diagnosis when it is evident, even if not specifically queried on the request form (and, assuredly, would soon be seeking the assistance of our medical defence organizations if we did not). Histopathologists have few qualms about diagnosing other ‘serious communicable diseases’—tuberculosis and hepatitis B and C—and, indeed, a granulomatous inflammatory picture incites a request for a Ziehl–Neelsen stain in an almost Pavlovian fashion, quite disproportionate to the likelihood of a ‘reward’ (positive result). We diagnose sexually transmitted infections, too, e.g. herpes simplex virus and Trichomonas vaginalis in cervical cytology samples, in the latter example seemingly unabashed by a hefty false-positive rate of up to 30%.63 Furthermore, we have a vehemently promulgated national cytopathology-based screening programme, targeting apparently healthy, asymptomatic young women, for a potentially fatal disease (carcinoma of the cervix) that is in most cases caused by a sexually transmitted organism, i.e. human papilloma virus. The 2008 UK National Guidelines for HIV Testing recommend that HIV serological testing be performed on all patients presenting for healthcare where HIV infection enters the differential diagnosis, and in all new general practice and general medical patients in high-prevalence areas.58 A logical conclusion drawn from this and the GMC and RCPath guidelines quoted above is that, in areas such as London, the informed consent procedure for all patients from whom samples are obtained for histopathological diagnosis should include the information that the histopathological diagnosis may reveal HIV infection. Every diagnostic histopathologist will know the unworkability and fallacy of restricting ‘testing’ to solely the diseases queried on the specimen request form, which are determined by the knowledge base, clinical acumen, speciality, diligence and handwriting size of, and time available to, the individual clinician who completes the form. We find it difficult to comprehend how a specimen received accompanied by ‘Lymphadenopathy? cause’ requires anything but the pathologist’s sincere and best efforts to reach a diagnosis, whatever that may turn out to be. This is surely better than concluding in a report that HIV infection should be excluded as a cause, leading to unnecessary discussions and requests for consent for HIV testing, for a patient whose only misfortune is to have one of the many morphological mimics of HIV disease. Until such time as this policy on informed consent is adopted nationally, it is our opinion that it is the histopathologist’s duty to do whatever is necessary with the specimen to fulfil the objective of sending the specimen to them, i.e. for a full and complete diagnosis. We are not advocating the indiscriminate use of p24 immunohistochemistry any more than we would condone (on financial and intellectual grounds) the unselected application of any other immunostain, but rather, would like to see the criteria for p24 immunohistochemistry usage brought into line with other antibodies, i.e. its judicious application in cases where there is clinical or histopathological morphological suspicion of HIV disease. We would advocate that every patient newly diagnosed as likely to have HIV disease by a histopathologist on the basis of morphology in conjunction with positive p24 immunohistochemistry should always then undergo formal HIV testing on peripheral venous blood instituted by the clinical team, in accordance with national guidelines.55 With the advent of HAART, histopathologists can perform HIV p24 immunohistochemistry confident in the knowledge that their actions accord with the first tenet of the GMC’s Duties of a doctor:60‘Make the care of your patient your first concern’. By thus treating HIV disease on a par with other diseases, histopathologists can add their voices to the burgeoning opinion in medicine, that the destigmatization and normalization of HIV disease are long overdue. In May 2009 the HIV Medicine Association of the Infectious of and the American College of published a on HIV which a that, routine screening for HIV infection and patients to be to

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.