Algorithms developed to facilitate the diagnosis of acute cutaneous porphyria

Two new algorithms for accurately diagnosing acute and cutaneous porphyria have been developed and validated, reports a Belgian study.

According to the study authors, these algorithms can be used to help clinicians correctly interpret lab tests related to porphyria.

“To our knowledge, this is the first time that diagnostic algorithms for acute and cutaneous porphyria have been developed and validated with sensitivity and specificity analysis,” the researchers write.

The study, “Development and validation of diagnostic algorithms for the laboratory diagnosis of porphyriaswas published in the Journal of Inherited Metabolic Diseases.

recommended reading

Porphyria is an umbrella term for a group of disorders in which heme precursor molecules and porphyrins build up and damage different tissues and organs. Heme is a molecule necessary for the transport of oxygen in living cells.

Several genetic and environmental factors are thought to trigger porphyria in susceptible people.

There are two major groups of porphyrias: cutaneous porphyrias and acute porphyrias. While cutaneous porphyrias primarily affect the skin, acute porphyrias are marked by sudden and potentially severe attacks that typically affect multiple organs and systems in the body.

Acute porphyrias include ALAD deficiency porphyria (ADP) and acute intermittent porphyria (AIP). Types of cutaneous porphyria include porphyria cutanea tarda (PCT), erythropoietic protoporphyria (EPP), X-linked erythropoietic protoporphyria (XLEPP), hepatoerythropoietic porphyria (HEP), and congenital erythropoietic porphyria (CEP).

People with hereditary coproporphyria (HCP) and variegate porphyria (VP) may experience acute attacks, as in acute porphyrias, or have skin symptoms.

A diagnosis is highly dependent on blood, urine, stool, and genetic testing. However, many clinicians have difficulty selecting and interpreting laboratory tests when making a diagnosis.

Diagnostic aid algorithms

Scientists from the University Hospitals of Leuven (UZ Leuven) and the Katholieke Universiteit (KU Leuven), Belgium, sought to develop algorithms to better diagnose acute and cutaneous porphyrias.

Patients with porphyria were identified in the UZ Leuven patient database and data from porphyria-related laboratory tests ordered between January 2000 and September 2020 were retrieved.

The research team ultimately included 639 patients who had clinical information. A total of 222 people were diagnosed with porphyria, while the remaining 417 were diagnosed with other conditions. The researchers developed two algorithms, one for acute porphyria and another for cutaneous porphyria, using information from a literature search, patient data and expert opinion.

The acute porphyria algorithm begins with the evaluation of heme precursor molecules, PBG and dALA, and porphyrins in urine. Acute porphyria is ruled out if patients test negative on all three tests while showing symptoms. Plasma and fecal porphyrins, as well as PBG deaminase activity, which is low in people with AIP, are also tested.

In the cutaneous porphyria algorithm, plasma and urine porphyrins are assessed when the patient has bullous lesions. If the patient has acute and painful photosensitivity, assays of plasma porphyrins and total erythrocyte protoporphyrins are performed.

Statistical analyzes determined the sensitivity and specificity of the algorithms. In this study, sensitivity (or true positive rate) refers to the proportion of people with acute or cutaneous porphyria who actually have the disease. Specificity (also known as the true negative rate) was used as an indicator of the algorithms’ ability to identify those who did not have the disease.

The sensitivity of the algorithm for acute porphyria was 100%. Thirteen cases of AIP and one of VP were identified. Its specificity was 98.5%.

For cutaneous porphyria, the sensitivity of the algorithm was 100% with a specificity of 93.9%. Seven cases of VP, 59 of PCT, 23 of EPP and two of XLEPP were identified.

The algorithms were also validated against cases from the European Porphyria Network (EPNET) External Quality Program. A total of 18 out of 19 cases of EPNET porphyria were correctly identified by the algorithms.

“The strength of our study is that the algorithms were validated using patient data. All algorithms available in the literature are built based on literature search and expert opinion, but there is no mention of validating the algorithms with clinical patient samples,” the researchers wrote. . “Both algorithms showed high sensitivity and specificity and can be used to help the physician correctly interpret lab results of porphyria-related tests.”

The researchers noted that there were no available diagnoses of HCP, ADP, CEP, or HEP, meaning that the sensitivity of the algorithm for these types of porphyria could not be validated.

Sharon D. Cole