Porokeratosis is one of the most common genodermatoses and is associated with an increased risk of keratinocyte cancer and melanoma

Abstract

Background

Porokeratosis is a clinically heterogeneous group of keratinization disorders with a genetic background mainly affecting the mevalonate pathway, which is involved in the synthesis of cholesterol, an essential component for the formation of the extracellular lipid lamellae in the stratum corneum. Porokeratosis is reportedly associated with an increased risk of keratinocyte cancer, but to date, no large epidemiological studies have been conducted to further address this association.

Objectives

The first objective was to characterize a cohort of patients diagnosed with porokeratosis at the Department of Dermatology and Venereology, Sahlgrenska University Hospital (SU), Gothenburg, Sweden. The second objective was to conduct a nationwide registry‐based cohort study to investigate the association, if any, between porokeratosis and the cutaneous malignancies squamous cell carcinoma (SCC), basal cell carcinoma (BCC) and melanoma.

Methods

For the SU cohort, the hospital registry was searched for patients with a diagnosis of porokeratosis recorded between 2016 and 2020. Clinical data were extracted from the records of the identified patients. For the nationwide cohort, national registries were searched to identify patients with a diagnosis of porokeratosis between 2001 and 2020. A tenfold control cohort was formed by Statistics Sweden. The data was cross‐referenced with the Swedish Cancer Register to study the associations between porokeratosis and SCC, BCC and melanoma.

Results

Disseminated superficial actinic porokeratosis was the most common clinical type among the 108 patients in the SU cohort. In the nationwide search, 2277 patients with porokeratosis were identified (prevalence 1/4132). Porokeratosis was associated with an increased risk for SCC, BCC and melanoma with hazard ratios (95% CI) of 4.3 (3.4–5.4), 2.42 (1.97–2.98) and 1.83 (1.18–2.82), respectively, in the patient cohort, compared to the matched control group.

Conclusion

Porokeratosis is a common genodermatosis, and it is associated with an enhanced risk of skin cancer.

INTRODUCTION

Porokeratosis is a clinically heterogenous group of keratinization disorders characterized by circular or annular skin lesions with a hyperkeratotic rim called the cornoid lamella. ¹ Although it is usually classified as a keratinization disorder, this disease could be reclassified as a genodermatosis, since recent studies have revealed that its underlying cause is a heterozygous germline mutation, usually of the genes that encode the enzymes of the mevalonate pathway. ² Case reports, case series and review articles since the 1970's have associated porokeratosis with malignancy, specifically cutaneous squamous cell carcinomas (cSSCs), ³ but, to date, substantial epidemiological studies on porokeratosis and the risk of skin cancer have not been available.

Vittorio Mibelli was the first to describe the classical form of porokeratosis in 1893. ⁴ Porokeratosis of Mibelli develops usually in children and begins with one or a few annular lesions with a raised border which slowly spreads centrifugally. ⁵ In addition to this classical form, several other clinical variants can be distinguished, e.g., disseminated superficial actinic porokeratosis (DSAP), disseminated superficial porokeratosis (DSP), linear porokeratosis, porokeratosis palmaris et plantaris disseminata, punctate porokeratosis, porokeratosis ptychotropica, genitogluteal porokeratosis and eruptive disseminated porokeratosis (Figure 1). ⁶ All clinical types exhibit a varying number of similar lesions but have a variable age of onset and bodily distribution. The most common type of porokeratosis is DSAP which occurs especially on the extremities. A similar but less common clinical variant, DSP, shows plaques in both sun‐exposed and unexposed areas. ⁷ Interestingly, current genetic data suggest that distinctions among these variants merely represent different phenotypes of the same disorder.

FIGURE 1.

Clinical characteristics of various types of porokeratosis. (a) a typical round porokeratosis lesion with a thicker keratinized rim (cornoid lamella) at the margin of the lesion on sun‐damaged skin. (b) Dermoscopic image of porokeratosis with a keratinzed rim. (c) a giant porokeratosis of Mibelli on the leg of patient: A reddish scaly, atrophic patch surrounded by a well‐defined ridge‐like border. (d) Disseminated superficial actinic porokeratosis (DSAP) lesions with multiple reddish brown, scaly spots on the extremities. (e) Female aged 31 years with unilateral linear porokeratosis since childhood. Lesions on her left leg before CO₂ laser treatment. (f) Same patient after CO₂ laser treatment.

Recent studies have identified heterozygous germline mutations of the mevalonate pathway genes in patients with various types of porokeratosis. ⁸ , ⁹ , ¹⁰ The mevalonate pathway produces sterols and isoprenoid metabolites and is a part of cholesterol biosynthesis. Cholesterol is an essential component of cell membranes and is needed for the formation of the extracellular lipid lamellae in the stratum corneum. ⁸ In addition, germline mutations in the solute carrier family 17 (SLC17A9) gene were identified in patients with familial DSAP. ¹¹ Altogether, over 200 pathogenic variants of the enzymes of the mevalonate pathway and of other selected genes have been registered in ClinVar and the Human Gene Mutation Database (Table S1).

Heterozygous germline mutations of the mevalonate pathway do not seem to cause significant structural or functional abnormality in the epidermis. However, each porokeratosis lesion exhibits individual second‐hit gene alterations in the wild type allele of the corresponding gene in epidermis. ¹² In DSAP, second hits differ among individual lesions, but in linear porokeratosis, second hits are identical in all individual lesions, which suggests that a single postzygotic somatic mutation is related to linear porokeratosis. Various factors predisposing to porokeratosis or triggering the phenotypic change have been suggested, such as ultraviolet (UV) radiation, immunosuppressive therapies ⁷ and human papilloma viruses. ¹³ These factors probably promote second hits and predispose at the same time to carcinogenesis.

In this study, we describe the clinical characteristics of a cohort of patients with porokeratosis identified in our department and validate the diagnoses by cross‐referencing them in the National Board of Health and Welfares Patient Registry. The nationwide Swedish register was further used to search for all patients in Sweden with porokeratosis from 1 January 2001 to 31 December 2022. The patient's risk for skin cancer was analysed by cross‐referencing with nationwide data recorded in the Cancer Registry and Basal Cell Cancer Register.

MATERIALS AND METHODS

The register study was conducted with the approval of the Swedish Ethical Review Authority (Approval number 2020‐06202).

Collection of clinical data

The digital patient registry of SU, Gothenburg, Sweden, was searched for the time between 2016 and 2020 for patients who had a doctor's appointment at the Department of Dermatology and Venereology and a diagnosis of porokeratosis (Q82.8T), according to the Swedish edition of the International Classification of Diseases, Tenth Revision (ICD‐10SE). The patient journals were scrutinized for clinical data, including family history, clinical and dermoscopic images, histology and molecular genetic analyses. The clinical and dermoscopic images were independently evaluated by three specialists in dermatology (RI, DK and SPE).

Using the unique personal identity codes, which are assigned to all citizens with permanent residence in Sweden, the local patient cohort was linked to the National Board of Health and Welfare's Patient Register. This register collects automatically information on all outpatient visits and inpatient treatments in public specialist care guided by the personal identification code and the ICD‐10SE diagnosis. To validate the porokeratosis diagnoses, we retrieved the diagnoses registered for these patients and calculated the positive predictive values (PPV) for porokeratosis diagnoses in the Patient Register.

Nationwide register data

The nationwide patient cohort was ascertained by searches of the inpatient and outpatient registers of the National Board of Health and Welfare for diagnosis Q82.8T (porokeratosis [Mibelli]) between 2001 and 2021. For calculation of the incidence and prevalence of porokeratosis, the population data of Statistics Sweden was used.

For calculating the risk for skin cancer of patients with porokeratosis, the patient and control groups (Table 1) were cross referenced with the National Cancer Registry, which included the Swedish Basal Cell Carcinoma (BCC) Register. Dates for porokeratosis were available from 1 January 2001, to 31 December 2019. Patients with porokeratosis were included in the cohort when they received a first diagnosis of porokeratosis. Only cases with a first diagnosis of porokeratosis (index date) recorded between 1 January 2006, and 31 December 2015, were included in the analysis of the results. For each case, 10 controls matched for age, sex and geographical location from the general population were randomly selected at the index date. Cases and controls with a history of a specific incident of cancer were excluded from the analysis of that type of skin cancer. All cases and controls were required to reside in Sweden from 1 January 2001, to the index date to be included. Patients were censored after the index date at their first emigration event, date of death or at the end of the observation period, which was 31 December 2019.

TABLE 1.

Demographic features of the nationwide porokeratosis and control groups analysed for keratinocyte cancers and melanoma

	Gender	Age			SD	n	N.A.	SEM	95% CI
		Min	Max	Mean
cSCC
Case	Male	13.0	97.0	67.6	14.1	247	0	0.90	65.9	69.4
	Female	11.4	95.6	65.9	13.7	676	0	0.53	64.9	67.0
	Total	11.4	97.0	66.4	13.8	923	0	0.46	65.5	67.3
Control	Male	13.0	95.6	67.9	14.6	2821	0	0.28	67.4	68.4
	Female	6.1	96.7	66.8	13.7	7613	0	0.16	66.5	67.1
	Total	6.1	96.7	67.1	13.9	10,434	0	0.14	66.8	67.3
BCC
Case	Male	13.0	91.3	67.9	13.8	242	0	0.89	66.1	69.6
	Female	11.4	95.6	66.2	13.8	667	0	0.53	65.2	67.3
	Total	11.4	95.6	66.7	13.8	909	0	0.46	65.8	67.6
Control	Male	13.0	97.0	67.7	14.7	2745	0	0.28	67.2	68.3
	Female	6.1	96.7	66.5	13.7	7366	0	0.16	66.2	66.8
	Total	6.1	97.0	66.9	14.0	10,111	0	0.14	66.6	67.1
BCC + cSCC
Case	Male	13.0	91.3	67.2	14.0	218	0	0.95	65.3	69.1
	Female	11.4	95.6	65.3	13.8	601	0	0.56	64.2	66.4
	Total	11.4	95.6	65.8	13.9	819	0	0.48	64.9	66.8
Control	Male	13.1	95.6	67.5	14.7	2705	0	0.28	67.0	68.1
	Female	6.11	96.7	66.4	13.7	7287	0	0.16	66.1	66.7
	Total	6.1	96.7	66.7	14.0	9992	0	0.14	66.4	67.0
Melanoma (including in situ)
Case	Male	13.0	95.6	68.6	14.0	274	0	0.85	66.9	70.2
	Female	11.4	96.7	66.8	13.8	748	0	0.51	65.8	67.8
	Total	11.4	96.7	67.3	13.9	1022	0	0.43	66.4	68.1
Control	Male	13.0	97.0	68.2	14.7	2854	0	0.27	67.6	68.7
	Female	6.1	96.7	66.9	13.7	7664	0	0.16	66.6	67.2
	Total	6.1	97.0	67.2	14.0	10,518	0	0.14	67.0	67.5

Abbreviations: BCC, basal cell carcinoma; CI, confidence interval; cSCC, cutaneous squamous cell carcinoma; N.A., not available; SD, standard deviation; SEM, standard error of mean.

Statistics

Kaplan–Meier plots were made covering the time from the index date to the first diagnosis of the respective type of skin cancer: BCC, cSCC (including cutaneous squamous cell carcinoma in situ), keratinocyte carcinoma (i.e., BCC and cSCC combined) and melanoma (including melanoma in situ). Hazard Ratios (HRs) were calculated using Cox's proportional‐hazards regression. All p‐values were two sided, and p < 0.05 was considered as statistically significant.

RESULTS

Local cohort

A total of 117 patients who visited our department in 2016–2020 had received a diagnosis of porokeratosis. The patient journals were reviewed to verify the diagnoses. Most patients were diagnosed clinically (Figure 2). A histopathological diagnosis was available for 36 patients, and nine patients were excluded from further analyses because of a histopathological diagnosis other than porokeratosis. Each excluded patient had a differential diagnosis typical for porokeratosis: benign lichenoid keratosis, pityriasis lichenoides chronica, nummular dermatitis, annular atrophic lichen planus, atrophic sun‐damaged skin, lichenoid dermatitis, squamous cell cancer in situ and two patients had actinic keratosis. Thirteen patients had been diagnosed through teledermatology, i.e., a clinical image from primary health care had been examined by a specialist. Photos of 20 clinically diagnosed patients were retrieved from the patient records and were evaluated by three dermatologists (RI, DK and SPE). The photos were in accordance with porokeratosis, and all patients were included in the study.

FIGURE 2.

Diagnostic methods to form the local porokeratosis patient cohort in the Sahlgrenska University Hospital, Gothenburg, in 2016–2020.

Of the 108 patients with a verified diagnosis of porokeratosis, the majority (n = 81, 75%) were women. The median age was 66 years. Five different clinical phenotypes were recorded: DSAP, DSP, porokeratosis of Mibelli, linear porokeratosis and genitogluteal porokeratosis (Table 2). DSAP was the most common phenotype. The most frequent location of porokeratosis was in the lower extremities (n = 68, 63%; Table 3). Only two patients had a family history of porokeratosis, but no genotyping had been carried out. Seventeen patients (15.7%) had undergone organ transplantation or used immunosuppressive medicines. In the patient cohort, 29 had BCC, three cSCC in situ, eight had cSCC, two melanomas in situ, seven had invasive melanoma and one had Merkel cell carcinoma.

TABLE 2.

Clinical characteristics of patients with porokeratosis in Sahlgrenska University Hospital

Porokeratosis type	Total, n (%)	Female, n (%)	Male, n (%)	Age	Median age	Transplantation and immunosuppression, n (%)
DSAP	56 (51.7)	44 (40.7)	12 (11.1)	31–94	69	7 (6.5)
DSP	1 (0.9)	1 (0.9)		15	15
Porokeratosis of Mibelli	44 (40.7)	34 (31.5)	10 (9.2)	34–89	65	8 (7.4)
Linear porokeratosis	1 (0.9)	1 (0.9)		26	26
Genitogluteal porokeratosis	2 (1.8)		2 (1.8)	22–70	46	1 (0.9)
N.A.	4 (3.7)	1 (0.9)	3 (2.8)	62–79	71	1 (0.9)
Total	108 (100)	81 (75)	27 (25)	15–94	66	17 (15.7)

Abbreviations: DSAP, disseminated superficial actinic porokeratosis; DSP, disseminated superficial porokeratosis; N.A., data not available.

TABLE 3.

Localization of porokeratosis in 108 patients

Body site	All cases, n (%)	DSAP, n (%)	DSP, n (%)	Mibelli, n (%)	Linear, n (%)	Genitogluteal, n (%)	N.A.
Upper extremities	11 (10.2)	3 (2.8)	1 (0.9)	7 (6.5)
Lower extremities	68 (63)	32 (29.6)		34 (31.5)			2 (1.8)
Both upper and lower extremities	18 (16.6)	18 (16.6)
Trunk	6 (5.5)	2 (1.8)		3 (2.8)			1 (0.9)
Genitogluteal	2 (1.8)					2
Head and neck	2 (1.8)				1 (0.9) a		1 (0.9)
N.A.	1 (0.9)	1 (0.9)
Total	108 (100)	56 (51.8)	1 (0.9)	44 (40.7)	1 (0.9)	2 (1.8)	4 (3.7)

Abbreviation: N.A., data not available.

^a Also, trunk and lower extremity.

When the porokeratosis diagnoses in the local cohort were compared to the diagnoses registered for these patients in the nationwide Patient Register, the PPV for porokeratosis was 93.1% (95% CI 86.2%–96.8%).

Incidence, prevalence, and cancer risk

The search of the nationwide Patient Registry identified 2277 patients in Sweden with at least one diagnosis of porokeratosis between 2001 and 2020. The patients were 5–98 years old at the time of inclusion, and their median age was 68 years. Most of the patients were female (71%, n = 1616). A total of 4929 outpatient or inpatient visits had been registered for porokeratosis, which yielded an average of 2.16 visits per patient. Most of the patients (67%) had made only one visit, and 90% had less than five visits. The incidence of porokeratosis was 1.2/100,000 person years, and the prevalence was 24.2/100,000, or 1/4132.

Cross‐referencing of nationwide porokeratosis and control populations with the National Cancer Registry showed an increased number of keratinocyte cancers in patients with porokeratosis compared with the tenfold control population (Figure 3). HRs (95% CI) in the patient cohort compared to the tenfold matched control group were 4.3 (3.4–5.4) for cSCC; 2.42 (2.0–3.0) for BCC; 2.8 (2.3–3.4) for keratinocyte cancer (patients having both cSCC and BCC) and 1.8 (1.2–2.8) for melanoma (Table 4).

FIGURE 3.

Kaplan–Meier curves showing cases of BCC (a), cSCC (b), both cSCC and BCC (c), or melanoma (d). Patients with porokeratosis are marked in blue and control persons in red line.

TABLE 4.

Hazard ratios for BCC and cSCC calculated as cases/controls

	Events		Hazard ratio	95% CI		Pr(>|z|)
	Individuals with porokeratosis, n	Controls, n
cSCC	101	271	4.30	3.42	5.40	p < 0.0001
BCC	108	506	2.42	1.97	2.98	p < 0.0001
BCC + cSCC	147	667	2.80	2.34	3.35	p < 0.0001
Melanoma (incl. in situ)	24	133	1.83	1.18	2.82	0.0065

Abbreviation: BCC, basal cell carcinoma; cSCC, cutaneous squamous cell carcinoma.

DISCUSSION

This large, register‐based study shows that porokeratosis is one of the most common genodermatoses and that there is an association between porokeratosis and skin cancer.

The collection of this cohort was made possible by the detailed ICD classification of genodermatoses in Sweden. Correct coding in patient records and administrative databases are essential for defining patient populations, planning and monitoring healthcare services and for conducting scientific research. However, many rare diseases do not have specifically assigned diagnostic codes, which is an obstacle to register studies. The WHO ICD‐10 contains no specific code for porokeratosis, whereas the Swedish version of ICD‐10, the ICD‐10SE has one code, Q82.8T, for porokeratosis of Mibelli. The scrutiny of medical records in our department revealed that all porokeratosis have been coded under porokeratosis of Mibelli, which is not incorrect because the various clinical porokeratosis types share a similar genetic background. The presence of this code enabled us to form both local and nationwide patient cohorts of porokeratosis.

Five different clinical types of porokeratosis were presented in the medical records obtained from our department. The DSAP was the most common clinical phenotype, which explains a relatively high median age of patients, 66 years. Only two patients had a family history of porokeratosis. It is likely that a family history had not been taken of elderly patients with DSAP, or maybe the patients were not aware of the symptoms in their family. None of the patients had been genotyped for porokeratosis. Once additional genetic data accumulates, the division of porokeratosis into several clinical types might need to be questioned.

A comparison of the porokeratosis diagnoses in the local clinical cohort and in the Patient Registry showed a high PPV, which suggests that the porokeratosis diagnoses registered in the nationwide data are sufficiently reliable for further study. Based on the proportions of the various clinical types of porokeratosis in the local data, DSAP is probably the most common clinical type also nationwide, as also suggested by the high median age of the patients recorded in the National Board of Health and Welfares Patient Registry. ¹⁴ DSAP has been associated with sun exposure, but according to our data and previous publications, DSAP is very uncommon in the face, the most sun‐exposed skin area of the human body. One can speculate that the face gets more chronic sun exposure which is known to induce thickening of the stratum corneum and therefore to protect from sun damage, while acute high‐dose exposure to the extremities promotes disruption of the stratum corneum and leads to second hits to the porokeratosis genes. The role of UV exposure on the pathogenesis of porokeratosis is, therefore, not completely clear, and a more appropriate name for disseminated superficial actinic porokeratosis (DSAP) could be late onset porokeratosis.

To our knowledge, this is the first epidemiological study which demonstrates that porokeratosis is associated with an increased risk of skin cancer. The follow‐up of patients with cSCC and BCC showed significantly more cancer cases among patients with porokeratosis than controls. The risk that the same person had both cSCC and BBC was also increased. Previous case reports and case series from single centers have reported that about 7%–11% of patients with porokeratosis develop cSCC or BCC. ¹⁵ , ¹⁶ , ¹⁷ Malignant transformation occurs reportedly in 19% of linear porokeratosis, in 7.6% of porokeratosis of Mibelli, and in 3.4% of DSAP. ¹⁸ A recent and largest study to date showed that 64 out of 110 patients with histologically confirmed porokeratosis had the history of skin cancer. ³ However, it was not possible to verify whether the cancer had arisen in the porokeratosis lesion in all cases.

Our results also showed a slightly increased risk for melanoma. Melanoma has previously been reported in three patients with porokeratosis: one in a case series of 11 patients with porokeratosis ¹⁶ and two in case reports which showed one superficial melanoma with a cornoid lamella at the edge of the melanocytic tumour and one amelanotic melanoma arising within a DSAP lesion. ¹⁹ , ²⁰

The pathophysiology of the association between cancer and porokeratosis is unknown. A genetic background of porokeratosis has only recently been demonstrated, ¹⁰ and it is becoming clear that the various clinical types of porokeratosis have a common molecular genetic basis, mainly in the genes encoding for enzymes of the mevalonate pathway. The mevalonate pathway has important functions in several cell processes, in addition to cholesterol synthesis. It produces, for example, steroid hormones, ubiquinones, and isoprenylated proteins such as RAS proteins. It is possible that the signalling which leads to carcinogenesis is mediated via MAPK and phosphoinositide 3‐kinase (PI3K)‐AKT signalling pathways or p53, which are known drivers of cSCC and melanoma. ²¹ , ²² , ²³

The main limitation of our study is the relatively small patient population in the local cohort whose clinical characteristics were retrieved from patient records. Risk factors for skin cancer, i.e., skin type, UV exposure, and family history of skin cancer, could not be obtained due to the retrospective study design. Nor was the porokeratosis diagnosis based on DNA analysis. Nine out of 117 patients (7.7%) in the local cohort had a false diagnosis based on histology. It is possible that the same number of incorrect diagnoses are present in the nationwide cohort, which may somewhat falsely decrease the reported skin cancer risk.

The Patient Register collects data from specialized health care. This may cause ascertainment bias towards patients with cancers because patients with mild forms of porokeratosis or those without cancer may not have visited specialized health care. Considering that patients with porokeratosis almost exclusively receive their diagnoses at dermatology departments, there is a risk for detection bias for skin cancer in this patient group, meaning that the HRs for skin cancer might be inflated. Another limitation of the study was that the nationwide data could only be delivered pseudonymized, and it was thus not possible to validate the diagnoses with clinical data. This study approach does not allow follow‐up of single porokeratosis lesions and cannot report figures on the risk of cancer of a porokeratosis lesion. The results of this study are, nevertheless, important to consider when recommendations for the follow‐up and management of porokeratosis are set.

In summary, porokeratosis is a relatively common multigenic genodermatosis with various clinical presentations. We have identified robust evidence that patients with porokeratosis have an increased risk of developing BCC, cSCC and, probably, melanoma. Further studies are warranted to identify associations between phenotype and genotype of porokeratosis and the mechanisms of carcinogenesis.

FUNDING INFORMATION

University of Gothenburg, Region Västra Götaland, Sahlgrenska University Hospital, Hudfonden.

CONFLICT OF INTEREST

None reported.

Supporting information

Table S1

Inci R, Zagoras T, Kantere D, Holmström P, Gillstedt M, Polesie S, et al. Porokeratosis is one of the most common genodermatoses and is associated with an increased risk of keratinocyte cancer and melanoma. J Eur Acad Dermatol Venereol. 2023;37:420–427. 10.1111/jdv.18587

DATA AVAILABILITY STATEMENT

Data are available upon request to the corresponding author, although data access is restricted.