Immune thrombocytopenia in the elderly: immunosenescent and clinical diversity.

Abstract

Immune thrombocytopenia (ITP) is recognised as the most common autoimmune cytopenic disorder and has been defined as a process associated with accelerated platelet destruction and decreased platelet production. Demographically, ITP affects adults and children with a bimodal distribution, involving cohorts aged <18 and 75–84 years.1 The most consequential clinical manifestation of ITP is its increased risk of bleeding; however, sparse data exist to characterise this risk at the older extreme of the ageing spectrum. When diagnosed in adulthood, most patients will eventually develop chronic disease and require long-term treatment strategies. The prevalence of ITP in adults is estimated to be 12 per 100 000 adults in the United States and data from the French National Health Insurance System has cited a prevalence of chronic ITP up to nine per 100 000 person-years in men aged >75 years.2, 3 More recently, data from the Nordic Country Patient Registry for Romiplostim estimated the prevalence of chronic ITP as 10–10·7 per 100 000 adults in Denmark and Sweden;4 however, we believe these are likely underestimates. In their paper Sokal et al.5 present data from the Cytopénies Auto-immunes: Registre Midi-PyréneEN (CARMEN)-France registry that expand and our understanding of the real-world presentation, risk factors for bleeding, and management of primary ITP in elderly patients (EPs), defined as those aged 65–79 years, as compared with the very EPs (VEPs) aged ≥80 years. The results complement the findings from other large ITP registries that have focussed on EPs and VEPs6, 7 and confirm the expectation that VEPs have a greater number of comorbidities and experience more severe bleeding events and higher mortality compared to younger cohorts. Interestingly, findings were similar between VEP and EP groups as far as the need for and response to first-line treatment, receipt of second-line treatment modalities, development of persistent disease, and complications of ITP and its treatment, including any bleeding occurrence, infection, and thrombosis. First-line therapy in VEPs more frequently utilised a combination of corticosteroids and intravenous immunoglobulin (IVIg) as opposed to single-agent corticosteroids. Factors associated with bleeding in VEPs included female sex, platelet count of <20 × 109/l and exposure to anticoagulants. Despite the insights provided by the Sokal et al.5 study and other large ITP registries that have examined EPs and VEPs, significant gaps in our knowledge persist. We lack prospective randomised controlled treatment trials for these age cohorts and recognise that elderly adults were typically excluded from clinical registration trials, due to age-related entry criteria, medical comorbidities, and, most important, selection bias exercised by investigators. Thus, we are relegated to analyse and apply the results generated from robust observational, registry-based studies to track the natural history of ITP and treatment responses in various subcohorts. In the study by Sokal et al.,5 which purports to describe the prospective real-world experience of primary ITP in France between 2013 and 2018, 251 of the 541 patients enrolled in the CARMEN-France registry were EPs or VEPs. However, an additional selection occurred such that 67 patients were subsequently excluded from the study group. The size of this excluded group (26·6%) is significant when compared with 97 in the EP group and 87 in the VEP group. A further analysis of these patients might identify potential pitfalls in the diagnosis and management of primary versus secondary ITP in the elderly. Of particular interest is whether their exclusion was related to bone marrow pathology or cytogenetic/mutational abnormalities. If their ITP was deemed secondary rather than primary this would have implications for how we should approach the diagnosis and follow-up of EPs and VEPs. Guidelines for the evaluation of isolated thrombocytopenia and ruling out secondary causes of ITP are not standardised and, as a primary example, the routine performance of bone marrow aspiration and biopsy is controversial. In the Italian Hematology Centers Registry,6 bone marrow biopsy was performed in only half of the studied patients. Per the International Working Group guidelines, a bone marrow biopsy is not recommended when the thrombocytopenia is isolated and there are no abnormalities on physical examination or on the peripheral blood film.8 And yet, age is a risk factor for clonal cytopenias and neoplasms. Studies indicate that 12% of myelodysplastic syndrome (MDS) may present with isolated thrombocytopenia of <100 × 109/l,9 the most common cytogenetic profile being normal. Up to 80% of secondary MDS have abnormal karyotypes, while <50% of patients with de novo MDS will have an abnormal karyocyte.10 Thus, next-generation sequencing on bone marrow aspirates may become a useful tool in EPs and VEPs with suspected ITP. Published clinical trials have typically considered the diagnosis of primary ITP to represent a ‘single’ disease entity, characterised by a ‘common’ thrombocytopenic phenotype. On the other hand, it is increasingly apparent that ITP is actually a heterogeneous disorder with disparate underlying pathophysiology and immunological triggers, which change with normal ageing and become exaggerated with extreme ageing. We lack ITP clinical trials that examine baseline and post-treatment changes in immunological repertoires; however, given the increased number of medical co-morbidities and changes to the immune milieu with age, it would not be surprising if nearly all primary ITP in EPs and VEPs results from secondary immunological phenomena. As observed by Cines et al.,11 primary ITP will comprise an ever-decreasing proportion of patients as specific inciting events and immune defects are better identified. Currently, we lack specific assays to define immunological dysfunction and immunosenescence, validated predictive biomarkers, or knowledge of their prevalence in different patient subpopulations. Nor do we have a complete understanding of the coagulation profiles in EPs with ITP that give rise to bleeding or, more rarely, thrombosis. The propensity for haemorrhagic or thrombotic complications is further confounded by concurrent comorbidities and antiplatelet and systemic anticoagulation strategies prevalent in EP and VEP cohorts. This issue speaks directly to the safety of various treatment modalities and how treatment decisions will be made in the EP and VEP. The currently available clinical guidelines do not specifically address treatment of ITP in EPs or VEPs, suggesting that the recommendations can be extrapolated across all adult age cohorts.8, 12 The lack of high-quality, prospective efficacy and safety data specific to EPs remains a major challenge when making clinical treatment decisions and clinicians must balance risk of harm with the promise of response. For example, corticosteroids remain an important front-line ITP therapy. Sokal et al.5 report higher than expected response rates to corticosteroids ± IVIg in VEPs and EPs. Yet, corticosteroids can incite or exacerbate problems already associated with ageing, including hypertension, hyperglycaemia, mood disturbances and agitation, and gastrointestinal (GI) bleeding. VEPs were particularly vulnerable to GI bleeding in this study (five of nine of the most serious bleeding events). Another knowledge gap involves the safety and efficacy in VEPs and EPs of the thrombopoietin receptor agonists (TPO-RAs). As observed in the French registry, and consistent with evolving guidelines in the management of ITP, TPO-RAs were the preferred second-line agent. However, age >60 years was a risk factor for venous thromboembolism (VTE) and VTE-associated mortality. The TPO-RAs have been associated with excess arterial thromboses, which is an important consideration in VEPs and EPs with underlying atrial fibrillation and peripheral arterial disease. The Italian Registry for elderly ITP reported the incidence of thromboses during TPO-RA treatment to be 3·6 per 100 patient-years, with a significant incidence of progression or recurrence, even in patients with severe thrombocytopenia or receiving concurrent antiplatelet/anticoagulant therapy.13 Clinical registration drug trials with the various TPO-RAs have included very few EPs and even fewer if any VEPs. There are no comparative safety or efficacy trial data among the TPO-RAs for any age groups but there are provocative data suggesting that the TPO-RAs may influence immune modulation by increasing regulatory T-cell activity and by decreasing release of certain inflammatory cytokines.14 This may be pertinent in EPs and VEPs, in whom dysfunction of regulatory T cells and dysregulation of immune homeostasis are involved in the development of different autoimmune diseases in old age.15 As we move into the future of treating patients with ITP of all ages, the immunoheterogeneity of this disease entity will need to be considered, particularly in EPs and VEPs. Ideally, these different profiles could lead to immune stratification in adequately powered randomised controlled trials in EP- and VEP-related ITP and may provide predictive biomarkers for safety and efficacy of new therapies. The success of this approach will require international collaboration. In this spirit, the Sokal et al.5 paper can serve as a ‘call to action’ for the creation of a single longitudinal, international registry, employing standardised definitions of diagnosis and safety and efficacy, and this effort should be combined with a biorepository to study the immunoregulation of ITP in all age cohorts. This aspirational endeavour offers the potential to understand real-world primary ITP versus ‘masked’ secondary ITP in the elderly.