X‐linked protoporphyria: Iron supplementation improves protoporphyrin overload, liver damage and anaemia

C-terminal deletions in the 5-aminolaevulinic acid-synthase 2 (ALAS2) gene were recently identified leading to the reclassification of a minority of patients with erythropoietic protoporphyria (EPP, Online Mendelian Inheritance in Man (OMIM) reference 177000) as having X-linked protoporphyria (XLPP, OMIM 300752). EPP and XLPP share clinical and biochemical features, such as photosensitivity, liver damage and elevated protoporphyrin IX (PPIX) (Whatley et al, 2008). A nine-year-old Caucasian boy presented with a seven-year history of sunlight-sensitive skin lesions and elevation of liver enzymes. EPP was diagnosed, based on excessive elevation of free erythrocyte PPIX (128 μmol/l, normal <0·09) and zinc protoporphyrin IX (znPPIX 1763 μmol/mol haem, normal <40). Further blood analysis showed microcytic anaemia (MCV 74·4 fl, normal >80; haemoglobin 114 g/l, normal >120) with iron deficiency and low ferritin concentration (5·9 μg/l, normal 12–63), iron (2·2 μmol/l, normal 4–24), transferrin (3·8 g/l, normal 2–3·6), transferrin saturation (2%, normal 17–42) and a raised soluble transferrin receptor (1085 ku/l, normal <710). As the patient showed no signs of malabsorption or malnutrition, intestinal function was considered normal. Liver enzyme activities were mild-moderately increased [alanine aminotransferase (ALT) 115·2 u/l, aspartate aminotransferase (AST) 61·8 u/l, gamma glutamyltransferase 93·6 u/l, normal <44, <50, <0·31, respectively). A liver biopsy demonstrated precirrhotic stage 3 with broad septa (Fig 1A, left) and spherical structures displaying red birefringence under polarized light with a dark ‘Maltese Cross’ in the centre (Fig 1A, right). Detection of a C-terminal deletion in ALAS2 (c.1706-1709 delAGTG) led to the diagnosis of XLPP (Whatley et al, 2008). Analysis of the FECH gene detected only insignificant polymorphisms. Treatment with oral beta-carotene, cholestyramine, tocopherol and ursodeoxycholic acid was started. Sunlight sensitivity improved and PPIX concentrations decreased within six months to 63·0 μmol/l. Based on a reported PPIX reduction of up to 85% and reversal of liver damage after successful allogeneic haematopoietic stem cell transplantation (HSCT) (Wahlin et al, 2007), the boy was selected for this procedure. During evaluation haemoglobin levels fell to 84 g/l, while liver enzymes (ALT 594 u/l and AST 426 u/l) and PPIX (94·6 μmol/l) increased. Metabolic liver decompensation was confirmed by a rise in bilirubin (36 μmol/l, normal <17) and a pathological International Normalized Ratio (1·8, normal 1) despite vitamin K supplementation. Treatment over the next 2 months consisted of 12 units red blood cells [RBCs; 12 units (1800 mg iron)], intravenous human haemin [400 mg haem arginate containing (103 mg iron)], and plasma- and erythrocyte-apheresis (Anstey & Hift, 2007). This resulted in an improved liver function that was paralleled by a more then 40-fold decrease of PPIX (2·0 μmol/l) (Fig 1B) and normalized body iron stores and serum ferritin (159 μg/l). Allogenic HSCT was performed in December 2007. Engraftment failure developed within 22 days leading to retransplantation using the patient's autologous stem cells. A further 17 RBC transfusions (2550 mg iron) were needed; follow-up revealed highly elevated ferritin (6090 μg/l), iron (39·9 μmol/l) and transferrin saturation (100%). Liver biopsy at this time demonstrated decreased protoporphyrin accumulation but siderosis of hepatocytes. Iron depletion was started using i.v. deferoxamin 25 mg/kg/day as infusion for 1 month and continued with oral deferasirox for further 16 months, with stepwise increases up to 25 mg/kg/day. To suppress the patient's erythropoiesis via feedback inhibition, erythrocytes were transfused when the haemoglobin fell below 129 g/l [12 concentrates within 12 months (1800 mg iron)]. After stopping RBC transfusions ferritin decreased to 467 μg/l and PPIX slightly increased to 9·6 μmol/l in the following 12 months. Subsequently, PPIX again increased up to 17 μmol/l and returned to < 10 μmol/l in response to oral iron supplementation without red cell transfusion. During the next 3 years, only iron replenishment produced tolerable PPIX concentrations ranging between 20 and 40 μmol/l, with only minor sunlight sensitivity, normal liver enzymes and improved liver biopsy findings. Iron substitution has been considered contraindicated or at least unpredictable in ferrochelatase-deficient EPP, because iron therapy (orally or via erythrocyte/haem-transfusions) has been reported to both worsen and improve EPP (Gordeuk et al, 1986; Milligan et al, 1988; McClements et al, 1990; Holme et al, 2007). Improvements may have occurred in patients with previously undiagnosed XLPP. RBC transfusions (total 6250 mg iron over 1·5 years) strikingly decreased PPIX blood concentrations (Fig 1B) in our patient and reduced protoporphyrin accumulation in a follow-up liver biopsy. The reasons and mechanisms for why many subjects with protoporphyria are iron-deficient remain unknown and is an important topic for further investigation. Recently it was shown that absorption of orally administered iron is normal in EEP/XLPP and that levels of hepcidin are appropriately low in these disorders (Bossi et al, 2014). We hypothesize that iron drives conversion of newly synthesized PPIX into haem, which can then be further metabolized. In erythroid precursor cells iron up-regulates, whereas iron deficiency down-regulates, ALAS2 expression via iron responsive elements and iron regulatory proteins. Apparently, with the C-terminal deletion within the ALAS2 gene resulting in a gain of function in XLPP, iron deficiency does not adequately down-regulate the rate limiting enzyme ALAS2 leading to excess production of PPIX and znPPIX (Whatley et al, 2008). On-going overproduction of PPIX may lead to iron consumption because ferrochelatase is an enzyme with high capacity that is not affected in XLPP (Puy et al, 2010). A lack of free iron obviously aggravates PPIX accumulation, initiating a phase of metabolic decompensation, which would explain why therapeutic iron can shift the disease back to the compensated phase (Fig 2). We believe that our findings significantly enhance our understanding of the pathogenesis and improve therapeutic management of XLPP. By converting toxic PPIX into haem, intravenously or orally administered iron sequesters PPIX and reverses serological and histological liver damage and photosensitivity. Therefore, replenishing depleted iron stores could be a simple and effective therapy for XLPP that probably obviates the need for HSCT and may prevent progression to cirrhosis. We thank Herbert Bonkovsky, MD, for valuable comments and Jacqueline Uhle for helpful assistance. The histological pictures were kindly provided by Dr. J.O. Habeck. CL for manuscript preparation and editing, data acquisition and statistical analysis, clinical care; KK for manuscript editing and revision, data acquisition and interpretation, clinical care; BG for clinical care, data interpretation, TS for porphyrin analysis, data collection and interpretation, manuscript editing and revision; SR for clinical care, data collection; DS for manuscript editing and revision; SW for mutation analysis, manuscript reviewing; JFB for manuscript editing, reviewing and clinical care; US for manuscript preparation and editing, data interpretation, clinical care, study coordination. The Authors declare, that there is no financial interest or possible conflict to report.