Portal de Periódicos da CAPES

Guidelines for the diagnosis and management of adult aplastic anaemia

2015; Wiley; Volume: 172; Issue: 2 Linguagem: Inglês

10.1111/bjh.13853

1365-2141

Sally Killick, Nick Bown, Jamie Cavenagh, Inderjeet Dokal, Theodora Foukaneli, Anita Hill, Peter Hillmen, Robin Ireland, Austin Kulasekararaj, Ghulam Mufti, John A. Snowden, Sujith Samarasinghe, Anna Wood, Judith Marsh,

Prenatal Screening and Diagnostics

The guideline group was selected to be representative of UK-based aplastic anaemia (AA) medical experts. Recommendations are based on review of the literature using MEDLINE and PUBMED up to December 2014 under the heading: 'aplastic anemia'. The Grading of Recommendations Assessment, Development and Evaluation (GRADE) nomenclature was used to evaluate levels of evidence and to assess the strength of recommendations. The GRADE criteria are specified in the BCSH guidance pack http://www.bcshguidelines.com/BCSH_PROCESS/EVIDENCE_LEVELS_AND_GRADES_OF_RECOMMENDATION/43_GRADE.html and the GRADE working group website http://www.gradeworkinggroup.org The objective of this guideline is to provide healthcare professionals with clear guidance on the management of patients with AA. The guidance may not be appropriate to every patient and in all cases individual patient circumstances may dictate an alternative approach. Review of the manuscript was performed by the British Committee for Standards in Haematology (BCSH) Haemato-Oncology Task Force, BCSH Executive Committee and then reviewed by a sounding board of the British Society for Haematology (BSH). This compromises 50 or more members of the BSH who have reviewed this guidance and commented on its content and applicability in the UK setting. It has also been reviewed by the Aplastic Anaemia Trust patient group but they do not necessarily approve or endorse the contents. Aplastic anaemia is a rare and heterogeneous disorder. It is defined as pancytopenia with a hypocellular bone marrow in the absence of an abnormal infiltrate or marrow fibrosis. To diagnose AA there must be at least two of the following (Camitta et al, 1975) haemoglobin concentration (Hb) <100 g/l, platelet count <50 × 109/l, neutrophil count <1·5 × 109/l. The majority (70-80%) of cases are idiopathic (Marsh et al, 2009). The remainder mainly consist of IBMFS. The incidence is 2-3 per million per year in Europe, but higher in East Asia (Montane et al, 2008). There is a biphasic distribution, with peaks at 10-25 years and over 60 years. AA not fulfilling the criteria for SAA or VSAA Patients commonly present with symptoms of anaemia and thrombocytopenia. Serious infection is not a frequent symptom early in the course of the disease. A preceding history of jaundice may suggest a post-hepatitic AA. Whilst the majority of cases are idiopathic, a careful drug, occupational exposure and family history should be obtained. Any putative drugs should be discontinued and the patient should not be re-challenged. If a possible drug association is suspected, this must be reported to the Medicines and MHRA using the Yellow Card Scheme (http://yellowcard.gov.uk). There is usually no hepatosplenomegaly or lymphadenopathy (except in infection). In young adults the presence of short stature, skin hyper/hypo pigmented areas and skeletal abnormalities, particularly affecting the thumb is suggestive of FA (Shimamura & Alter, 2010). The triad of nail dystrophy, reticular skin pigmentation and oral leucoplakia is characteristic of dyskeratosis congenita (DC) (Shimamura & Alter, 2010). The finding of peripheral lymphoedema may indicate a diagnosis of Emberger syndrome due to germline GATA2 mutation. See Table 1 for the summary of investigations for the diagnosis and further evaluation of AA; this table also summarizes the emerging diagnostics incorporating the latest molecular technologies that are likely to feature in the diagnosis and differential diagnosis within the next couple of years. Both a bone marrow aspirate and trephine biopsy are required for the diagnosis of AA, and the key bone marrow findings are summarized in Table 2. Karyotyping may fail in very hypocellular marrows with there being insufficient metaphases. In this situation perform FISH analysis for chromosomes 5, 7, 8 and 13 It was previously assumed that the presence of an abnormal cytogenetic clone indicated a diagnosis of MDS and not AA. However it is now evident that abnormal cytogenetic clones [such as del(13q), trisomy 8 and others], which may be transient, are present in up to 12% of patients with otherwise typical AA at diagnosis (Gupta et al, 2006; Afable et al, 2011b). Although monosomy 7 may indicate the likelihood of MDS in children, in adults monosomy 7 can also be seen in AA. Abnormal cytogenetic clones may arise during the course of the disease and the appearance of a new cytogenetic abnormality may provide evidence of clonal evolution (Maciejewski et al, 2002) The investigations in Table 1 should exclude non-AA causes of pancytopenia with a hypocellular bone marrow, which are listed in Table 3. A MDT meeting approach is recommended to collate relevant results and develop a treatment plan. Consideration should be given for seeking expert advice on the diagnosis and management of patients where there is uncertainty, or when an IBMFS is being considered. A number of inherited/genetic disorders are characterized by BMF/AA, usually in association with one or more somatic abnormality (Alter, 2007). The BMF typically presents in childhood but this can sometimes be in adulthood. The two syndromes frequently associated with generalized BMF/AA are FA and DC (Dokal, 2011; Soulier, 2011), which can sometimes present with AA alone as their initial manifestation. These syndromes are genetically heterogeneous; 16 FA genes and 10 DC genes have been identified. The FA genes are important in DNA repair, the DC genes in telomere maintenance. Based on the DNA repair defect a diagnostic test-'chromosomal breakage test' is available for FA. Patients with DC usually have very short telomeres and this measurement [using flow cytometric fluorescence in situ hybridization or multiplex quantitative polymerase chain reaction (PCR)] can be useful in the assessment of DC. Genetic testing for known DC genes (representing c. 60% of cases) is possible in specialized centres. In addition there are other genetic syndromes that are sometimes associated with AA/cytopenias. This includes Shwachman‒Diamond syndrome ‒ SDS (Dror et al, 2011) (mutations in SBDS), congenital amegakaryocytic thrombocytopenia ‒ CAMT (Ballmaier & Germeshausen, 2011) (mutations in MPL) and GATA2 deficiency (Emberger syndrome) (Horwitz, 2014) as well as genetically uncharacterized cases. Some cases of inherited AA first present in adulthood and it is important to recognize these as their management differs from that of idiopathic AA. Where there are sufficient characteristic abnormalities a diagnosis may be straightforward (e.g. mucocutaneous features in DC). Where the presentation is only with AA and with minimal non-haematological abnormalities, inherited BMF should be considered and testing for known BMF syndromes should be undertaken. Investigations for inherited forms of AA should be re-appraised in patients initially classified as "idiopathic AA" and who fail to respond to anti-thymocyte globulin (ATG). For most patients with AA, transfusion with red blood cells (RBC) is essential to maintain a safe blood count, improve symptoms of anaemia and maintain quality of life. The decision to transfuse RBC should be based on clinical symptoms (signs of anaemia), taking into consideration the patient's age and co-morbidities (cardiac, pulmonary or vascular). Although no specific pre-transfusion haemoglobin concentration (Hb) trigger can be recommended, it is important to maintain quality of life and avoid symptoms. A higher trigger may be needed for elderly patients and those with co-morbidities. Optimal use of RBC transfusion involves administration of enough red cells to maximize clinical outcome whilst avoiding unnecessary transfusions (Carson et al, 2012). Alloimmunization against red cell antigens and iron overload are the commonest risks associated with regular transfusion therapy. Provision of phenotype-matched blood (for Rh and Kell) should be considered to reduce the risk of alloimmunization. Regular platelet transfusion support may be required for AA patients. With the exception of one publication (Sagmeister et al, 1999), literature specific to platelet transfusion support in AA is lacking, and evidence is taken from studies addressing the need for platelet transfusion support in patients with reversible thrombocytopenia (Estcourt et al, 2012; Stanworth et al, 2013; Killick et al, 2014). It is recommended that prophylactic platelet transfusions should be given to stable AA patients on active therapy (where the treatment aims to reverse the severe thrombocytopenia) with a platelet count 20 × 109/l. For thrombocytopenic patients requiring invasive procedures, platelet transfusions must be administered, aiming to achieve a platelet count in line with BCSH guidelines for the relevant procedures (British Committee for Standards in Haematology, 2003), and a pre-procedure platelet count should be checked. During treatment with ATG, worsening thrombocytopenia can occur. This is due to increased platelet consumption in the presence of cross-reacting antibodies in ATG binding to platelets. Although there are no studies to support the exact threshold for platelet transfusion support prior to ATG, most authors use a threshold of 20 × 109/l (Scheinberg et al, 2011; Scheinberg & Young, 2012). Regular support with RBC and platelet transfusions increases the risk of HLA and non-HLA (minor histocompatibility) alloimmunization, leading to poor platelet increments and increased risk of graft rejection after HSCT. Leucodepletion of cellular blood components may reduce, but not eliminate, alloimmunization (Killick et al, 1997; Desmarets et al, 2009). The possibility of HLA alloimmunization and provision of HLA-selected platelets should be considered for patients refractory to platelet transfusion, provided other causes of refractoriness have been excluded. In the absence of HLA antibodies and for patients failing to increment with HLA-matched platelets, investigation and matching for human platelet antigen antibodies should be considered. The use of irradiated granulocytes should be considered in patients with life-threatening infection related to severe neutropenia (Quillen et al, 2009), and anecdotally may be life saving. Data about the effectiveness of granulocyte concentrates are limited and usage is linked with a number of adverse events, such as transfusion-related acute lung injury, alloimmunization and febrile reactions. Irradiation of cellular blood components prevents transfusion-associated graft-versus-host disease (TA-GVHD). This is a rare complication of blood transfusion with 100% mortality. Irradiation may also reduce the risk of alloimmunization in AA, as reported from animal data (Bean et al, 1994). Following universal leucodepletion in the UK, the Advisory Committee on the Safety of Blood, Tissues and Organs (SaBTO) no longer recommends the use of cytomegalovirus (CMV)-negative blood components (if they have been leucodepleted) for patients with immunodeficiency (unless pregnant) and those undergoing HSCT (SaBTO Annual Report, 2011/12), although PCR monitoring should be considered (https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/215126/dh_132966.pdf). To date, there has not been a statement from the British Society of Bone Marrow Transplantation regarding blood products and CMV status. CMV-negative granulocyte components should be provided for CMV-negative recipients. Aplastic anaemia patients on regular RBC transfusion support will develop tissue iron overload, but there remains debate on the clinical impact of transfusional iron overload. In the setting of HSCT, a raised serum ferritin is an adverse predictor of outcome in myeloablative stem cell transplantation (Armand et al, 2007). Although unreliable, serum ferritin remains the most widely quoted parameter for assessment of iron overload. Magnetic resonance imaging (T2* or R2) can quantitate cardiac and liver iron, and is a useful adjunct although its utility in AA has not been published. There are few published data regarding iron chelation therapy in AA. A large study was the 1-year Evaluation of Patients' Iron Chelation with Exjade study (Lee et al, 2010). This confirmed that chelation with deferasirox can be administered safely in patients with AA (no drug-induced cytopenias were noted), and can reduce the serum ferritin. However, dose adjustments are required to adequately chelate those who are heavily transfusion dependent. Impaired renal function is observed with deferasirox, and the drug should be used with caution in AA patients who are taking CSA. Deferasirox is licensed for use in transfusion-dependent anaemia, but only as second line therapy when desferrioxamine is inadequate or contra-indicated. Deferiprone is efficacious but not recommended in neutropenic patients (Cermak et al, 2011). For those responding to immunosuppression, or after a successful HSCT, venesection is recommended for iron overload. Infections remain the major cause of death in AA (Marsh & Kulasekararaj, 2013). In contrast to cancer patients undergoing chemotherapy, in SAA neutropenia is prolonged and persistent, resulting in a higher incidence of invasive fungal infection (IFI) and severe bacterial sepsis. Survival of non-responders to ATG in the last two decades has markedly improved and this has occurred in conjunction with decreased infection-related mortality and decreased frequency of IFIs (Valdez et al, 2011). Aplastic anaemia patients who are severely neutropenic should ideally be nursed in isolation when in hospital. In the UK it is common practice to give prophylactic antibiotics and antifungals, regular mouth care including an antiseptic mouthwash (such as chlorhexidine or saline) and food of low bacterial content (Hochsmann et al, 2013). Prophylactic antibiotics, either two non-absorbables (e.g. colistin and neomycin) or quinolones (e.g. ciprofloxacin), may be initiated but the preference should be according to local policy. A mould (aspergillus) active azole, preferably itraconazole or posaconazole, should be used as prophylaxis. In the UK, prophylaxis against Pneumocystis jirovecii is not routinely given. Anti-viral prophylaxis in untreated patients with AA is not routinely given. Antiviral prophylaxis with aciclovir or valaciclovir should be used during and after ATG therapy. During ATG therapy, sub-clinical reactivation of CMV and Epstein–Barr virus (EBV) is common but self-limiting, and therefore does not need antiviral treatment; EBV-related post-transplant lymphoproliferative disease has only very rarely been reported after ATG, most often after rabbit ATG. It is not UK practice to give Pneumocystis jirovecii prophylaxis with ATG. Protocols and guidelines for the management of febrile neutropenia, including the assessment and management of fungal infections, are well developed and clinicians should follow local hospital and National Institute for Health and Care Excellence guidance (Phillips et al, 2012). Empirical anti-fungal therapy, as per local guidelines, should be initiated early for patients with clinically suspected IFIs, as these patients have persistent neutropenia. Granulocyte transfusions may be potentially life saving in severe sepsis, such as invasive fungal disease, particularly for patients due to proceed to HSCT (Quillen et al, 2009). Haemopoietic growth factors, such as erythropoiesis-stimulating agents and granulocyte colony-stimulating factor (G-CSF), are usually ineffective in supporting blood counts in AA patients (Marsh et al, 2007), although encouraging preliminary results are reported with the thrombopoietin-mimetic, eltrombopag (Desmond et al, 2014); see also section on Treatment of AA in the Elderly. Standard first line IST is the combination of horse ATG (ATGAM; Pfizer, New York, NY, USA) and CSA. Lymphoglobuline horse ATG is no longer available (Marsh et al, 2009; Passweg & Marsh, 2010; Scheinberg & Young, 2012). A prospective randomized study from the National Institutes of Health (NIH) and a prospective EBMT study showed significantly better response at 3 and 6 months, and survival with horse ATG compared to rabbit ATG for first line IST (Scheinberg et al, 2011; Marsh et al, 2012). There is no indication for routine use of G-CSF with ATG + CSA (Tichelli et al, 2011). Prednisolone is used with ATG for the sole purpose of prevention of side effects of ATG. There is no upper age limit for ATG, but there is increased mortality in patients aged >60 years treated with ATG (Tichelli et al, 1999, 2011) (see later section on Treatment of AA in the Elderly). A second course of ATG may be indicated for failure to respond or relapse after a first course or if the patient is ineligible for UD HSCT (Marsh et al, 2009; Passweg & Marsh, 2010; Scheinberg & Young, 2012) (see Fig 2). For a second course, rabbit ATG may be given. A second course of horse ATG is an alternative option, but this may be associated with more immediate and late (serum sickness) side effects (Marsh et al, 2012). Compared to horse ATG, rabbit ATG produces more profound and prolonged lymphodepletion and, in some recent studies, more infections. It is therefore important to ensure that patients receive adequate prophylactic antimicrobial support when using rabbit ATG. The dose of horse ATG (ATGAM) is 40 mg/kg/d for 4 d. It is given as an intravenous infusion over 12-18 h. Due to the risk of anaphylaxis, a 'test' dose must be given. Current practice is to use an intravenous infusion test dose (recent survey of the EBMT SAA Working Party, unpublished data May 2012), whereby the first 100 ml of the first day infusion is given over 1 h. ATG should be given through a double lumen Hickman or other central venous catheter, as it is sclerosing to peripheral veins, and also for ease of administration of other drugs and blood products. Each dose of ATG should be preceded with intravenous methyl prednisolone 1 mg/kg, chlorphenamine, and platelet transfusions aiming to keep the platelet count >20–30 × 109/l (Marsh et al, 2009; Scheinberg & Young, 2012). Broad-spectrum intravenous antibiotics according to local policy should be given for febrile episodes irrespective of the neutrophil count. Fluid retention occurs commonly during ATG treatment, especially in older patients; careful attention to fluid balance is important. Prednisolone is started on the day after ATG is completed at a dose of 1 mg/kg/d for 2 weeks, followed by rapid tapering over the 2 weeks. Ciclosporin should be commenced as the prednisolone dose is tapered, at a dose of 5 mg/kg/d to achieve trough blood levels of 100–200 μg/l. CSA should be continued whilst the blood count continues to rise. A slow tapering of the drug (25 mg every 2‒3 months) can be started after at least a further 12 months of therapy, to reduce the risk of later relapse (Dufour et al, 2013). Side effects of ATG are (i) early reactions, including fever, rash, rigors, hypo/hypertension, fluid retention, rarely acute pulmonary oedema/adult respiratory distress syndrome and anaphylaxis and (ii) later, serum sickness occurring days 7‒14 from the start of ATG, most commonly with arthralgia, myalgia, rash and fever. Serum sickness is treated with intravenous hydrocortisone 100 mg four times a day (QDS) and adequate analgesia; it usually requires a few days of treatment. Extra platelet transfusions are often needed during the period of serum sickness due to platelet consumption. There is no indication for using G-CSF with ATG + CSA, as prospective randomized studies have shown that daily G-CSF given for 3 months after ATG does not improve response or overall survival (OS) (Tichelli et al, 2011). Response to ATG (as defined in Table 4a,b) is delayed, starting after an average of 3‒4 months. The 6-month response rate to a first course of horse ATG is around 70%. Five-year OS is age-dependent: 100% for age 60 years (Tichelli et al, 2011). In comparison, the response to a first course of rabbit ATG is around only 35‒45%, with significantly worse OS (Scheinberg et al, 2011; Marsh et al, 2012; Scheinberg & Young, 2012). For NSAA, ATG + CSA results in significantly higher response rates, 74% versus 46% (and better event-free survival), compared to CSA alone (Marsh et al, 1999). Relapse after ATG occurs in up to 35% of patients; the risk of later clonal evolution to MDS/acute myeloid leukaemia is 15%, and haemolytic PNH in 10% (Rosenfeld et al, 2003; Scheinberg & Young, 2012). Transfusion independence (if previously dependent) or doubling or normalization of at least one cell line or increase of baseline Response to a second course of ATG from most studies is around 35% for refractory AA and 55–60% for relapsed AA (Marsh et al, 2009; Passweg & Marsh, 2010; Scheinberg & Young, 2012). Factors predicting for response are summarized in Table 5. It is recommended that expert advice be sought when considering the use of other immunosuppressive drugs. Mycophenolate mofetil, sirolimus, corticosteroids and cyclophosphamide are not recommended in the treatment of AA (see Table 6). There is a potential relapse risk of AA following vaccinations in those patients who have responded to IST. The evidence base is limited and based on anecdotal case reports, as well as an appreciation that a viral insult is likely to be an important trigger in the pathogenesis of AA (Viallard et al, 2000; Hendry et al, 2002). Vaccinations, including influenza vaccination, should be avoided if possible, except following HSCT, when AA patients should be routinely vaccinated as recommended for all allogeneic bone marrow transplantation recipients (see HSCT section). The current indications for HSCT are based on the EBMT SAAWP guidelines (Sureda et al, 2015). Patients should be managed in JACIE [Joint Accreditation Committee-International Society for Cellular Therapy (ISCT) and EBMT]-accredited centres. Up-front HSCT from a MSD is indicated for SAA in young and adult patients who have a MSD. EBMT data show similar outcomes for patients aged 40-50 to those aged 30-40 years (Sureda et al, 2015). However, co-morbidities should be carefully assessed to determine fitness for up-front transplantation instead of IST for patients aged 35-50 years. Unrelated donor HSCT is indicated for SAA after failure to respond to one course of IST. There is no strict upper age limit but this should be discussed on an individual patient basis and according to co-morbidities at the respective transplant centre. The donor should be 10/10- or 9/10-matched based on HLA high resolution typing for class I (HLA-A, -B, -C) and II (HLA-DRB1, -DQB1) antigens. Alternative donor HSCT using either cord blood, a haploidentical family donor or a 9/10-matched UD may be considered, among other treatment options, after failure to respond to IST and in the absence of a MSD and a suitably matched UD (Samarasinghe et al, 2012; Passweg & Aljurf, 2013). All donors should be screened for donor-directed HLA antibodies, the presence of which is associated with a very high risk of graft rejection. There is less clear guidance on the exact indication for alternative donor HSCT as this is less successful than MSD or UD HSCT, but new approaches to alternative donor HSCT are being evaluated using uniform EBMT protocols. In the rare situation where there is a syngeneic donor available, HSCT should be considered in all patients regardless of age as long term OS exceeds 90% (Marsh & Kulasekararaj, 2013). For all newly diagnosed AA patients who may be potential transplant candidates, HLA tissue typing should be performed at time of diagnosis, so that (i) MSD HSCT can proceed as soon as possible, and ideally before the patient becomes sensitized, not only to HLA but also to minor histocompatibility antigens, and (ii) the potential availability of UDs is established, so that if there is no response to a course of ATG and CSA, the patient can then proceed to UD HSCT (or earlier if the patient's condition is of concern with severe and/or recurrent infections). Assessment for response to IST is usually made at 3‒6 months. An MDT approach is essential for the pre-transplant work up. The aims of the work up are to (i) confirm the diagnosis and exclude/document clonal evolution (ii) assess co-morbidities (iii) select the donor, conditioning regimen, stem cell dose and source, (iv) address fertility issues and (v) inform the transfusion laboratory of the potential transplant and review of transfusion requirements (Table 7). The choice of conditioning regimens to use depends on (i) patient age (ii) type of donor (iii) centre preference for choice of antibody, whether ATG (Bacigalupo et al, 2010; Sanders et al, 2011) or alemtuzumab (Marsh et al, 2011; Bacigalupo et al, 2012). See Table 8. For adult MSD HSCT, the survival is age-dependent, but OS is 70-85% between the ages of 30 and 50 years. A recent EBMT analysis has shown that outcomes after UD HSCT are no longer inferior to MSD HSCT, in that UD is not a negative predictor of survival (Bacigalupo et al, 2013; Marsh et al, 2014). Specific issues relating to AA HSCT regarding early post-transplant management and management of late effects are summarized in Table 9. The treatment of elderly patients (aged >60 years) with AA is more complex than in younger patients. In addition, the outcome is worse due to inferior tolerability of the treatment. Therefore patients should be individually assessed for co-morbidities and their specific wishes should be respected, as quality of life is an important outcome in this group. With regard to diagnosis, it is important to exclude hypoplastic MDS, as MDS is far more common than AA in this age group (see diagnostic section). Older age per se, is not a reason to withhold treatment even in the very elderly. Immunosuppression is considered the treatment of choice. There is no place for allogeneic HSCT as first line therapy in patients aged >60 years, although HSCT can be considered in selected patients with a syngeneic donor. Ideally, the least toxic and most convenient treatment should be given. However, another consideration is how quickly a response is required, such that those with life threatening cytopenias (neutrophil count <0·2 × 109/l) or having suffered a severe infection requiring hospitalization should be treated more intensely than those with less severe disease. Treatment with ATG and CSA results in a more rapid and complete response than CSA alone in patients with NSAA (Marsh et al, 1999). However, patients require hospitalization and have a higher risk of acute and delayed toxicity than younger patients, so the risks and benefits of treatment should be weighed up for each individual patient. Patients must be assessed carefully before treatment, as the risk of infection, bleeding, heart failure and arrhythmias with ATG is higher in the elderly. Older patients have an inferior survival after ATG compared to younger patients (Tichelli et al, 1999). Alternative treatments include CSA alone, oxymetholone (or danazol) or alemtuzumab. Although the response rate of CSA alone is inferior to the combination of ATG and CSA in NSAA, OS is not inferior as CSA-refractory patients may respond to second line therapy with ATG and CSA (Marsh et al, 1999). CSA alone has the convenience of being outpatient-based but patients must be carefully monitored for nephrotoxicity and hypertension. Alemtuzumab may be used as a single agent in refractory/relapsed AA, but medical fitness needs very careful assessment in older patients prior to considering this agent as a possible option (Scheinberg et al, 2012). Oxymetholone or danazol can be considered in men intolerant or unresponsive to CSA (Allen et al, 1968; Jaime-Perez et al, 2011). Danazol has fewer masculinizing side effects than oxymetholone so may be a better alternative for women. Careful monitoring of oxymetholone is required as it can cause nephrotoxicity, hepatic tumours, mood changes, cardiac failure, prostatic enlargement and raised blood lipids. Patients who are intolerant of, or who decline IST should be offered best supportive care. Eltrombopag is a peptide, small molecule, oral thrombopoietin receptor agonist. In an extension of an earlier phase II study at NIH, 43 patients with refractory SAA were treated with eltrombopag (Desmond et al, 2014). Haematological responses, including trilineage response, were observed in 40% of patients. The drug was well tolerated in most patients. Elevated transaminase levels may occur and there are particular concerns about clonal evolution, including monosomy 7, which requires further evaluation. Eltrombopag has been approved by the Food and Drug Administration in the USA for treatment of SAA refractory to IST. It has recently, as of August 2015, been licensed by the EMA for SAA refractory to IST or patients who are heavily pre-treated and unsuitable for HSCT. It should be used with meticulous long term monitoring for clonal evolution, or following a clinical research protocol. It is advised that a repeat bone marrow is performed prior to starting treatment to exclude an abnormal cytogenetic clone typical of MDS/AA, particularly monosomy 7. Although the relationship, either casual or coincidental, between AA and pregnancy is controversial, it remains a serious condition, challenging to manage and with a variable clinical outcome. AA can be diagnosed for the first time during pregnancy, in early or late gestation. Cytopenia(s) often progresses during pregnancy, but the disease may remit spontaneously, after abortion (spontaneous or therapeutic) or after delivery (Aitchison et al, 1989). Relapse is common during pregnancy in AA patients who have previously responded to ATG, especially those with partial response (Tichelli et al, 2002). Pregnancy does not trigger relapse of the disease in patients who had undergone successful HSCT. Tichelli et al (2002) evaluated outcomes among 36 pregnancies in women previously treated with immunosuppression for AA. They reported almost half involved a complication in the mother (three abortions, two cases each of eclampsia and maternal deaths) and/or baby (five premature deaths). Relapse of AA occurred in 19% and a further 14% needed transfusion during delivery. Normal blood counts before conception did not guarantee freedom from relapse of AA during pregnancy. Better supportive care in recent years, particularly in supply of blood products, has led to improvements in maternal and fetal outcome (Kwon et al, 2006). However, it is important to discuss with the patient and family the potentially serious risks to both the mother and baby (Deka et al, 2003). It is essential that the patient be monitored frequently throughout pregnancy, initially monthly but later more frequently and according to disease severity, and with very close liaison with the obstetric team and haematologist. Presence of a PNH clone should warrant discussion with a specialist centre. The mode of delivery should be determined on obstetric grounds. Supportive care is the mainstay of treatment of AA in pregnancy and the platelet count should, if possible, be maintained above 20 × 109/l with platelet transfusions. The high risk of alloimmunization and platelet refractoriness needs to be considered. CSA is safe during pregnancy (McKay & Josephson, 2006) and is recommended for those needing transfusions. ATG, allogeneic HSCT or androgens for AA during pregnancy are not recommended. Paroxysmal nocturnal haemoglobinuria should be excluded by performing flow cytometry (Parker et al, 2005; Borowitz et al, 2010). Analysis of GPI-anchored proteins is a sensitive and quantitative test for PNH, enabling the detection of small PNH clones which occur in up to 50% of AA patients, the proportion depending on the sensitivity of the flow cytometric analysis used (Dunn et al, 1999; Sugimori et al, 2006). Such small clones are most easily identified in the neutrophil and monocyte lineages in AA and will be detected by flow cytometry. If the patient has had a recent blood transfusion, a population of GPI-deficient red cells may still be detected by flow cytometry in the granulocyte and monocyte population. However, the clinical significance of a small PNH clone in AA as detected by flow cytometry remains uncertain. Such clones can remain stable, diminish in size, disappear or increase, hence the need for monitoring the clone. What is clinically important is the presence of a significant PNH clone often associated with clinical or laboratory evidence of haemolysis. Urine should be examined for haemosiderin as this is a constant feature of haemolytic PNH even when the patient does not have macroscopic haemoglobinuria. Evidence of haemolysis associated with PNH should be quantitated with the reticulocyte count, serum bilirubin, serum haptoglobin and lactate dehydrogenase. Patients should be screened for PNH at the diagnosis of AA. If persistently negative, test 6 monthly for 2 years and then move to annual testing unless symptoms/signs develop. If the PNH screen is, or becomes, positive, test 3-monthly for the first 2 years and only reduce the frequency if the proportion of the PNH cells has remained stable. The presence of a PNH clone in the setting of AA does not directly influence the choice of therapy for the underlying BMF. There is some evidence that the finding of a PNH clone predicts a better response to IST but this is not universal in all published reports. Patients with a significant PNH clone receiving IST, especially ATG, should be actively monitored for signs of haemolysis. Conversely, AA may later emerge in PNH patients in the presence of significant haemolysis. New PNH patients should be referred to one of the two specialized nationally commissioned PNH centres, St James's University Hospital, Leeds, and King's College Hospital, London, to be assessed for PNH complications and for consideration for anti-complement therapy, following formal PNH National Service MDT review. Patients will be seen in either of the two national centres or in one of 10 Outreach clinics. Data from the French Registry compared to the EBMT outcomes demonstrates that allogeneic stem cell transplant has an inferior outcome in haemolytic and thrombotic PNH compared to best supportive care including eculizumab when indicated (Peffault de Latour et al, 2012). Therefore the finding of a PNH clone does not affect positively or negatively on the decision to transplant. While the advice and information in these guidelines is believed to be true and accurate at the time of going to press, neither the authors, the British Committee for Standards in Haematology (BCSH) nor the publishers accept any legal responsibility for the content of these guidelines. These guidelines are only applicable to adult patients with AA. We would like to thank Professor Neal Young and Professor Jakob Passweg for their review of this document. The authors would like to thank the BCSH task force and the BSH sounding board, BCSH executive committee and the Aplastic Anemia Trust Patient Group for their support in preparing these guidelines. SBK chaired the guidelines group. JCWM was the senior author. All authors were involved in formulation, writing and approval of the guidelines. All authors approved the final version of the manuscript. All authors have made a full declaration of interests to the BCSH and Task Force Chairs, which may be reviewed on request. In summary the following authors have declared the following conflicts of interest: SBK has received payment from Celgene for speaking at education meetings and from Novartis for speaking at education meetings and advisory work; TF has received payment from Alexion; ID has received payment from Life Length for lecturing; AK has received funding from Alexion for speaking at educational meetings; JM has received funding from Pfizer, Sanofi, Novartis and Alexion; GM has received funding from Celgene; JS has received funding from Merck Sharp and Dohme, Celgene, Orthobiotec and Pfzier. PH, AH have received payment from Alexion for lecturing. The rest of the authors have no declarations of interest. Members of the writing group will inform the writing group Chair if any new pertinent evidence becomes available that would alter the strength of the recommendations made in this document or render it obsolete. The document will be archived and removed from the BCSH current guidelines website if it becomes obsolete. If new recommendations are made an addendum will be published on the BCSH guidelines website at www.bcshguidelines.com. If minor changes are required due to changes in level of evidence or significant additional evidence supporting current recommendations a new version of the current guidance will be issued on the BCSH website.