Skin Cancers: Causes, Diagnostic, and Modern Detection Tools

Nowadays, skin cancer is one of the most relevant diseases affecting fair-skinned populations, counting over 5 million of new victims every year. From understanding the molecular mechanisms of most occurring cancers, such as melanoma and keratinocyte cancers, to developing powerful and modern prevention and diagnostic tools, skin cancer research has brought important improvement in the management of this overwhelming disease.

Cancer remains one of the most prevalent diseases world-wide, accounting for 4.5 million deaths yearly and being one of the leading causes of premature death (i.e., at ages between 30-69 years). Among the many different discovered types, skin cancer, which refers to pathological entities that arise from cells in the epidermis or dermis, is the most common cancer affecting humans. It manifests predominantly in fair-skinned individuals, with an incidence increasing with age and solar radiation exposure (Wild et al., 2020). Up to 95% of skin cancers derive from melanocytes (melanoma) or from epidermally derived cells (non-melanoma skin cancers)(Craythorne & Al-Niami, 2017). With over 5 million new cases reported every year, skin cancer research is blooming, both in mechanistic understanding and prevention and diagnostic tools development (Viale, 2020).

Cancer basics: Neoplasm, benignancy, malignancy.

Neoplasm is the autonomous growth of tissues, acquired through a stepwise process derived from the accumulation of mutations in critical cancer genes. The two main types of cancer genes are proto-oncogenes, genes that promote cell growth, and tumor suppressor genes, genes that inhibit cell growth and induce cell death. Therefore, activating mutations acquired in proto-oncogenes lead to a gain of function, causing the affected cells to multiply uncontrollably, while deactivating mutations in tumor suppressor genes enable neoplastic cells to escape death and persist in the body (Barnhill et al., 2004). When a neoplasm develops a solid mass, we refer to this as a tumor. Tumors are classified as benign, when they remain localized, or as malignant or cancer, when they spread to distant sites in the body (Rubin et al., 2011).

Figure 1: Benign vs Malignant tumors. A benign neoplasm grows slowly and only locally and presents no metastases. A malignant neoplasm grows rapidly, infiltrates other tissues, and can metastasize (Whelan, 2022).

Benign Melanoma: Common Acquired Melanocytic Nevi (Moles)

During the first two years of life, most people develop 10 to 50 nevi on their skin, the number depending on light exposure or inherited susceptibility. The main cause of nevus formation is an activating mutation of the gene encoding the oncogene BRAF, leading to increased numbers of melanin pigment producing cells, melanocytes, in the basal epidermis. Naturally, with an enlarged number of these cells, hyperpigmentation can be observed appearing as a small tan dot, initially with a diameter of one to two millimeters. During the following three to four years, the dot grows to become a uniform brown circular area with a well-defined outline, with a diameter of up to four to five millimeters, that usually stops from any subsequent growth (Rubin et al., 2011).

Malignant Melanoma: the Most Aggressive Type of Skin Cancer

The incidence of malignant melanoma is increasing dramatically, reaching the estimation that more than 1% of the children born today will develop this affection. Melanoma has several subtypes and can arise either from UVB exposure, host factors, or the joint action of both. With increasing numbers or sizes, nevi pose the highest host factor risk of developing malignant melanomas. In some cases, nevi can acquire different combinations of mutational events, which can be identified by the apparition of dysplastic nevi, i.e., nevi that do not follow the normal growth pattern, presenting an especially strong risk of developing malignancy and eventually metastasis.

The prognosis of most melanomas is excellent when detected and treated early, but this deteriorates at an alarming rate with thicker lesions. The ABCDE criteria can be used as a checklist for melanoma suspicion. Subsequently, the disease is histologically diagnosed by analyzing an excision of the affected skin, which also determines the treatment. Early phase melanomas can be cured by wide local excision, while for metastatic melanomas, immunotherapy and targeted therapy provide the best prognosis (Craythorne & Al-Niami, 2017).

Figure 2: Warning signs of melanoma: Asymmetry, border, color, diameter, evolution. (ABCDE’s of Melanoma)

Non-melanoma skin cancers: Basal Cell Carcinoma and Squamous Cell Carcinoma

Keratinocyte cancers, basal cell carcinoma (BCC) and squamous cell carcinoma (SCC), are the most frequently occurring cancers. Even though they exhibit very low mortality rates, these cancers imply heavy financial impact on the health systems, due to the high abundance and inferred costs of diagnosis and treatment. Both BCC and SCC are strongly correlated with UVB and therefore appear mainly on the body parts with the most sun exposure (Wild et al., 2020).

BCCs tumors grow slowly but can be locally aggressive. The chance of metastasis is very low, but they pose the risk of recurring due to neglect or inappropriate treatments. BCCs arise from pluripotent cells within the follicular epithelium. On a genetic basis, BCC is mainly caused by a deactivating mutation in the PTCH tumor suppressor gene. Unlike SCC, BCC also occurs on skin that has little light exposure, as the PTCH mutation can be inheritable (Rubin et al., 2011). It can present characteristics such as pearly papules, non-healing scabs, patches displaying redness and slightly rolled edges, or thickened white areas with scar appearance.

SCCs are faster growing tumors arising from squamous keratinocytes in the skin or mucous membrane epidermis. Besides UVB, the second most prevalent factor causing these tumors is the Human Papilloma Virus (HPV). Although the risk is very low (<2%), SCCs can also metastasize. SCC symptoms can vary greatly from one case to another, ranging from crusted papules to ulcerated nodules or wart-like formations. Following a biopsy-based histological diagnostic, both BCC and SCC can be identified as low-risk or high-risk, for which surgical excision and radiotherapy are the most common treatments (Craythorne & Al-Niami, 2017).

Figure 3: (Left) BCC with a pearly edge (Craythorne & Al-Niami, 2017). (Right) Flat SCC with a scaly crust.

Skin Cancer: Primary & Secondary Prevention

As UVB is the main factor involved in all types of skin cancer, primary prevention involves encouraging behaviors that minimize hazardous sunlight exposure. As Green et al. have proved, applying sunscreen regularly significantly decreases the risk of developing melanoma. Secondary prevention refers to halting disease progression and improving outcomes by detecting and initiating treatment at an early stage. Thus, the main tools for secondary prevention are skin self-examination, which can be easily done at home, either individually or with the help of a family member, or periodic skin examinations by health care providers (Rojas et al., 2022). With the modern technology available to both doctors and average users, secondary screening became greatly accessible and could therefore be a strong weapon against the high incidence of skin cancers.

Artificial intelligence as a tool for skin cancer detection

The term Artificial Intelligence (AI) refers to developing intelligent machines that can carry tasks autonomously. Among the subfields of AI, we can count Machine Learning (ML), field that studies how computers can learn tasks automatically. By using convolutional neural networks (CNNs), a method of ML that specializes in learning the relationship between input (e.g., images) and classification tasks (e.g., diagnosis), great success has been achieved in dermoscopic image analysis. CNNs brought along important delevopments in early skin cancer detections, providing a strong tool that is both accessible to average users, by employing smartphone applications in self-examination, or to dermatological specialists who can now access automated diagnostic tests. Even though AI-based diagnostics come with limitations such as lack of patient and lesion clinical data and history, this is still an active area of research than can already facilitate the flow of cancer detection and that is prone to improve at fast rates (Reiter et al., 2019).

Conclusions

Among all types of cancer, skin cancer accounts for most victims yearly. Researchers have done outstanding work so that we can understand how notorious forms of skin cancer, such as melanoma, basal cell carcinoma, squamous cell carcinoma work, both on a molecular and pathological level. Even though prevention mechanisms are widely available, this matter keeps reoccurring, arising the need to develop strong detection and diagnostic methods, for which artificial intelligence has brought a significant turnover. Although currently limited, modern tools provide solid hope for the future of dermatological diseases.

Bibliographical References

Barnhill, R. L., Piepkorn, M., & Busam, K. J. (2004). Pathology of Melanocytic Nevi and Malignant Melanoma. Springer New York.https://doi.org/10.1007/978-0-387-21619-5

Craythrone, E., & Al-Niami, F. (2017). Skin cancer. Medicine. https://doi.org/10.1016/j.mpmed.2017.04.003

Green, A. C., Williams, G. M., Logan, V., & Strutton, G. M. (2011). Reduced melanoma after regular sunscreen use: Randomized trial follow-up. Journal of Clinical Oncology, 29(3), 257–263.https://doi.org/10.1200/JCO.2010.28.7078

Reiter, O., Rotemberg, V., Kose, K., & Halpern, A. C. (2019). Artificial Intelligence in Skin Cancer. Current Dermatology Reports, 8(3), 133–140. Current Medicine Group LLC 1. https://doi.org/10.1007/s13671-019-00267-0

Rojas, K. D., Perez, M. E., Marchetti, M. A., Nichols, A. J., Penedo, F. J., & Jaimes, N. (2022). Skin cancer: Primary, secondary, and tertiary prevention. Part II. Journal of the American Academy of Dermatology, 87(2), 271–288.https://doi.org/10.1016/j.jaad.2022.01.053

Rubin, R., Strayer, D. S., & Rubin, E. (2011). Rubin's Pathology: Clinicopathologic Foundations of Medicine, Sixth Edition. Lippincott Williams & Wilkins.

Viale, P. H. (2020). Cancer Facts & Figures. Journal of the Advanced Practitioner in Ocology, 11(2), 135-136. American Cancer Society. https://doi.org/10.6004/jadpro.2020.11.2.1

Wild, C. P., Weiderpass, E., & Stewart, B. W (2020). World Cancer Report: Cancer research for cancer prevention. World Cancer Report (pp. 374-381).

Visual Sources