Introduction

Hereditary porphyria is a metabolic disorder that develops when any of the enzymes involved in the heme synthesis system is genetically impaired. This disease significantly impairs the patients’ quality of life; currently, there is no radical treatment for hereditary porphyria indicating that patients may develop serious sequelae, which may be fatal. Erythropoietic porphyria (EPP) is caused by mutations in the FECH gene1. In 10%–20% of patients with EPP, hepatic dysfunction is observed due to deposition of erythrocyte protoporphyrin and serum protoporphyrin in hepatocytes and bile canaliculi2,3. Approximately 2%–5% of patients with EPP die because of liver damage gradually progressing to cirrhosis or liver failure (chronic liver failure), which rapidly develops into irreversible cholestatic liver failure (acute liver insufficiency). Regarding the onset of liver dysfunction, in cases with mutations in the FECH gene that lead to loss of enzyme activity, the precursor protoporphyrin potentially accumulates, leading to an increased incidence of liver damage. In contrast, while it is reported that aggravation of liver damage correlates with protoporphyrin levels4, other study have reported cases without any correlation between liver damage and serum protoporphyrin levels5. Currently, our understanding about the association of liver damage with EPP is limited.

Ursodeoxycholic acid6, cimetidine7,8,9, cholestyramine10, and plasmapheresis11, among other compounds, have been reported to be effective for liver damage in EPP. Moreover, liver transplants have been reported to be beneficial; however, the long-term clinical course after treatment remains unclear12. Therefore, there is an urgent need to identify the most effective treatment for patients with severe liver damage.

In this study, we report the efficacy of phlebotomy in EPP patients with severe liver injury treated at our hospital.

Results

Patient background

The median age at referral was 31 years (26–64 years). Of the seven patients, six were male and one was female. Family history was observed in 6 patients and photosensitivity was observed in all 7 patients. The serum PP level at the time of referral was as high at 4059 (range 2973–14,518) µg/dL red blood cells (RBCs) (Table 1). EPP was confirmed in six patients by genetic testing (Table 2)13. All the EPP patients in which FECH mutations were identified are predicted to be caused by the pathogenic FECH mutation in combination with the low expression allele c.315-48C in trans13.

Table 1 Patient background.
Full size table
Table 2 Genetic testing for EPP diagnosis.
Full size table

Liver function and iron metabolism test at the time of referral

Table 3 shows the results of each liver function test at the time of referral. Hepatocellular injury type was noted in one patient (No. 6), cholestatic type in two (No. 1, 4), mixed type in one patient (No. 5), and an unclassifiable type in three patients (No. 2, 3, and 7)14. In addition, two patients showed jaundice at the time of referral (No. 1, 3). No obvious anemia was observed in 7 patients at the time of referral.

Table 3 Liver function and iron metabolism at the first visit.
Full size table

Treatment and prognosis

Thorough shading was performed in all the nine patients. Ursodeoxycholic acid was used in four patients, cimetidine in five patients, and colestyramine in two patients (Table 1). Since these existing treatments were ineffective, 3 patients (No1, No5, No6) subsequently added phlebotomy. No cases of liver cancer were observed during the course of the study until March 2021.

Laboratory data immediately before phlebotomy and efficacy of phlebotomy

Blood chemistry data (collected immediately before phlebotomy) are summarized for the three patients (No. 1, 5, and 6) who underwent phlebotomy. All the three patients had overt jaundice and significantly elevated serum PP levels (Table 4).

Table 4 Laboratory data immediately before and after phlebotomy.
Full size table

In patient No. 1, plasma exchange was performed five times; although serum PP levels were decreased, hepatic injury did not improve. Phlebotomy was performed four times in total (400 mL was removed in the first time and 200 mL was removed thereafter). A decrease in the serum PP level and subsidence of liver injury was observed, along with a decrease in the Hb and serum ferritin levels (Fig. 1).

Figure 1
figure 1

Clinical course after phlebotomy. In patient No. 1, plasma exchange was not performed. Phlebotomy therapy was performed three times (first time, 400 mL; following 2–3 times, 200 mL). In patient No. 5, a plasma exchange was performed five times; however, the serum PP level and liver damage did not improve; thus, the treatment was changed to phlebotomy. Phlebotomy was carefully performed four times in total (400 mL the first time; 200 mL thereafter). In patient No. 6, plasma exchange was performed four times; however, improvement in the serum PP level and hepatitis is not obtained. Thus, the treatment was changed to phlebotomy. Phlebotomy was performed four times in total (400 mL the first time; 200 mL thereafter).

Full size image

In patient No. 5, plasma exchange was performed five times and the serum PP levels were decreased; however, liver injury did not improve. Phlebotomy was conducted four times in total (400 mL was removed in the first time and 200 mL removed thereafter), and a decrease in the serum PP level was observed with improvement in liver injury (Fig. 1).

In patient No. 6, phlebotomy therapy was performed three times (400 mL was removed in the first time and 200 mL was removed thereafter) from the beginning; the serum PP level increased and liver injury improved along with a decrease in the Hb and serum ferritin levels (Fig. 1).

Pathological findings

Liver biopsy was performed in 3 cases, and the Maltese cross was confirmed in all cases. In No. 5 and No. 6, PP deposits were distributed in the bile duct and hepatocytes. On the other hand, in No. 2, PP deposition was mainly in the bile duct and less was deposited in hepatocytes (Fig. 2). The stainability of ABCG2 was guaranteed in No. 2, On the other hand, it showed a marked decrease in No. 5 and No. 6 (Fig. 2).

Figure 2
figure 2

Hematoxylin and eosin (H&E) staining and ABCG2 staining pathology. Liver biopsy was performed in four patients, and the Maltese cross was confirmed in all patients. Biopsy samples were subjected to H&E and ABCG2 staining.

Full size image

Moreover, to evaluate damage in the cell membrane of affected hepatocytes, double staining with ABCG2 and cadherin was performed (Fig. 3). In patient No. 5, cadherin was well expressed in PP-deposited hepatocytes, whereas ABCG2 was stained, suggesting the selective loss of ABCG2 activity on the cell membrane. Although the expression of cadherin in PP-deposited hepatocytes was heterogeneously detected in patient No. 6, ABCG2 staining was reduced in hepatocytes with PP deposition, indicating the loss of ABCG2 activity on the preserved cell membrane as well as damage to the cell membrane. Contrastingly, in patient No. 2, the expression of cadherin and ABCG2 was maintained.

Figure 3
figure 3

ABCG2 and cadherin double staining pathology. To evaluate the function of the hepatic cell membrane, cadherin staining was performed and double staining with ABCG2 was performed. In No. 5, cadherin (red) was strongly stained in PP-deposited hepatocytes, but ABCG2 (green) was weakly stained (black circles). In No. 6, both the staining properties of ABCG2 (green) and cadherin (red) were decreased in PP-deposited hepatocytes (the part indicated by the red circle). On the other hand, in No. 2, the stainability of ABCG2 and cadherin was guaranteed.

Full size image

Discussion

In general, liver damage is a factor that mostly affects the prognosis of patients with EPP. Approximately 5–10% of all patients with EPP have liver damage, where 1% of these patients have fatal conditions2,3. Therefore, there is an urgent need to establish effective treatment for EPP patients with severe liver damage. EPP is caused by the accumulation of excess protoporphyrin in RBCs; hence, plasmapheresis may be useful for removing haemolyzed and spilled protoporphyrin in the blood. However, it was not effective in patient No. 5 and 6, suggesting that sufficient protoporphyrin could not be removed. Therefore, plasmapheresis was followed by phlebotomy in these patients15,16,17. In patient No. 1, phlebotomy was introduced from the beginning. Phlebotomy is theoretically suitable for the removal of protoporphyrin in erythrocytes. However, hypermyelination due to the progression of anemia caused by phlebotomy may induce further production of protoporphyrins, which may worsen the condition. Therefore, we carefully performed a 400-mL phlebotomy for the first time by monitoring the blood data (such as Hb levels) and then slowly performed a 200-mL phlebotomy once every 1–2 weeks. After phlebotomy, liver damage rapidly subsided simultaneously with the decrease in protoporphyrin in all three patients; liver injury relapse was not noted thereafter. Phlebotomy has been proposed for congenital erythropoiesis (Günter's disease), acute liver and cutaneous porphyria18. This strategy is expected to suppress heme biosynthesis through the regulation of ALAS, a pathway restriction enzyme that leads to the accumulation of porphyrins. Although the effectiveness of phlebotomy in EPP patients has been shown this time, it is necessary to investigate the mechanism such as the regulatory action of ALAS in the future.

There are also reports on the onset and exacerbation factors of liver damage in EPP patients.

It has been reported that PP has low water solubility, which is an important mechanism for the onset of liver damage; the excretion of PP into the bile duct causes inflammation due to viscosity19. Furthermore, it has been reported that PP level and liver damage are correlated4. However, there are many cases in which the PP level does not correlate with the severity of liver damage, and therefore, it is necessary to analyze the mechanism of onset and severity of liver damage. In this study, we focused on ABCG2, a type of hepatocyte transporter, because it is involved in porphyrin transport20,21,22. We have previously reported the development of different levels of liver damage between siblings with the same PP level and the association between ABCG2 staining and liver damage5.

In this study, we increased the number of patients and conducted additional studies. Liver biopsies were performed in 3 of the 7 patients (No. 2, 5, and 6) to examine ABCG2 expression. ABCG2 staining was lower in patients No. 5 and 6 who had more severe liver damage than in patients No. 2. However, we also considered the possibility that the decreased expression of ABCG2 was affected by reactive oxygen species production due to the accumulation of protoporphyrin in hepatocytes. To solve this problem, we performed double staining with cadherin, which is a tight junction. Patient No. 5 maintains cadherin expression in protoporphyrin-deposited hepatocytes, indicating a selective loss of ABCG2 in the conserved cell membrane of hepatocytes. From this result, the accumulation of porphyrins in hepatocytes with loss of ABCG2 may be important for the onset and exacerbation of liver damage. However, we have not directly proved the causal relationship between the decreased expression of ABCG2 and liver damage, and further investigation is required in the future.

In recent years, there have been very interesting reports that ABCG2 deficiency protects against EPP-related hepatotoxicity23. It is considered that one of the mechanisms of liver damage protection is that the excretion of PP into the bile duct is reduced due to the decreased expression of ABCG2. On the other hand, in actual human patients, PP accumulation in hepatocytes and apoptosis of hepatocytes with decreased ABCG2 expression can be observed, so there may be a mechanism different from that of the mouse model. In any case, further research is needed on mechanism analysis.

In conclusion, phlebotomy has proven to be an effective treatment option in EPP patients with severe liver damage. Further research is needed to elucidate the pathophysiology of EPP in order to suppress porphyrin production and improve liver damage.

Methods

Patient background

We examined seven patients (No. 1–7) who were diagnosed with EPP and liver damage.

Ethics declarations

Written informed consent were obtained from all of the patients enrolled in this study and ethical permission of this study was granted by the Review Boards of Kindai University Faculty of Medicine (approval number 25-085). All experiments were performed in accordance with the Declaration of Helsinki.

Pattern classification of liver damage

The pattern classification of liver damage was performed using the drug-induced liver damage diagnostic criteria (JDDW2004)14. Hepatocellular injury type; ALT > 2N + ALP ≤ N or ALT ratio/ALP ratio ≥ 5, cholestatic type; ALT ≤ N + ALP > 2 N or ALT ratio/ALP ratio ≤ 2, mixed type; ALT > 2 N + ALP > N and 2 < ALT ratio/ALP ratio < 5 (N: Upper limit of normal, ALT ratio = ALT value/N, ALP ratio = ALP value/N).

Phlebotomy

Phlebotomy was first performed by removing 400 mL and then by removing 200 mL several times, depending on the patient's condition.

Histological analysis

Percutaneous liver biopsy was performed in four patients. Liver biopsy samples were subjected to hematoxylin and eosin (H&E) staining. Immunohistochemical analyses were performed using mouse anti-human adenosine triphosphate-binding transporter G2 (ABCG2) antibody (Abcam, Cambridge, United Kingdom, BXP-21) and rabbit anti-human E-Cadherin (Cell Signaling Technology, Danvers, MA, 24E10). For fluorescence staining, secondary antibodies labeled with Alexa 488 or Alexa 555 were used (Life Technologies, Carlsbad, CA). Fluorescent images were taken using a confocal laser microscope (Carl Zeiss GmBH, Jena, Germany).