Introduction Erythropoietic protoporphyria and X-linked protoporphyria, collectively referred to as the protoporphyrias, result in accumulation of protoporphyrin IX, causing severe cutaneous pain and, in a minority of patients, liver failure. Approximately half of patients with protoporphyria have microcytic hypochromic anemia, which may be related to reduced iron availability or decreased heme biosynthesis. Several reports have described mild thrombocytopenia in patients with protoporphyria as well, though the mechanism for this remains unknown. In this study, we characterize the relationship between erythrocyte metal-free protoporphyrin IX (PPIX) levels, platelet count, and iron status in patients with protoporphyria.

Methods We analyzed hematologic parameters in patients with protoporphyria treated at the Mass General Brigham (MGB) hospital system. Patient-level data across multiple time points were collected. Generalized Estimating Equation (GEE) models with an identity link and Gaussian family distribution were performed to determine the association, accounting for within-participant associations. A separate analysis was performed using linear regression for a cohort of patients with protoporphyria in Sweden, with one value per patient.

Results A total of 118 patients with protoporphyria were included in the analysis including 65 patients from MGB and 53 from the Swedish cohort. In the MGB cohort, the median PPIX level was 1482 ug/dl (min 265- max 4897, ref <20). A 1mg/dl increase in PPIX was associated with a 0.02 x10⁹/L decrease in platelet count on average (p< 0.0001). This relationship was consistent when adjusting for alanine transaminase (ALT) and ferritin. There was no association between PPIX and ferritin (p = 0.57), PPIX and soluble transferrin receptor (p=0.31), or PPIX and hemoglobin (p =0.40). Higher PPIX levels were associated with lower MCV (p = 0.019) and higher ALT (p <0.0001). Similarly, in the Swedish cohort, a 1 mmol/L (ref <1.2) increase in PPIX was associated with a 1.4640 x10⁹/L decrease in platelet count (p = 0.01224), a relationship that remained consistent when adjusting for ALT and ferritin. PPIX was not significantly associated with hemoglobin (p=0.102), ferritin (p=0.352), MCV (p=0.063), or ALT (p=0.068), but the trends for MCV and ALT were present in the same direction as for the US cohort.

Conclusion In this study of 118 patients with protoporphyria in the US and Sweden, we demonstrated a linear relationship between PPIX and platelet count that remained consistent when controlling for ALT and ferritin. The PPIX level was not associated with hemoglobin or markers of iron stores. Moving forward, we plan to present data for matched controls in the MGB cohort as well as data on the iron regulatory hormones, hepcidin and erythroferrone, in this group of patients.