Advanced cutaneous squamous cell carcinoma – a pressing case for treatment

Non-melanoma skin cancers (NMSC) are the commonest cancers in Western countries, but the sheer size of the problem means that it has been difficult to capture the total burden of disease. Few national registers exist but estimates of the incidence in the US suggested an incidence of >5.4 million cases in 2012.1 A national database established in the UK has identified more than 210 000 cases in 2015 and suggests a year-by-year increase of >7%.2 Accurate data are critical for the provision of good skin cancer care pathways and service planning as well as arguing for significant investment in new treatments.3, 4 Non-melanoma skin cancers predominantly derive from epidermal keratinocytes, and so it has recently been suggested that they be renamed keratinocyte cancers (KC),5 although in fact keratinocytes from mucosal sites also give rise to cancers. Basal cell carcinomas form the largest group of KC in all existing registers, but a rising incidence in an ageing population has highlighted the significant morbidity and mortality of patients with advanced cutaneous squamous cell carcinoma (cSCC). There is therefore an urgent need for new innovative treatments for such patients, who have a poor prognosis and poor response to chemotherapy. Estimates of metastasis in cSCC vary from 0.5 to 3.3%.6 Within the UK national database, 1566 cases of metastatic cSCC were identified between 2013 and 2015, with an overall incidence of 1.4% (2.4% in men and 1% in women). The poor prognosis was highlighted by the finding that 93% of the metastases developed within 3 years (mean time to metastasis 342 days) with a 3-year overall survival of 49%.7 The groundbreaking finding of PTCH gene mutations underlying familial basal cell naevus syndrome8 subsequently identified Hedgehog signalling as the critical pathway in basal cell carcinogenesis and led to a new era of targeted therapy with the development of Hedgehog inhibitors.9 Although there are multiple genetic syndromes predisposing to cSCC10 which have elucidated some key biological mechanisms, such as DNA repair in xeroderma pigmentosum- and pigmentation-associated genes in albinism, these have not led on the whole to novel targeted therapies. The field is complicated in cSCC by a very high mutational burden (as with basal cell carcinoma (BCC) and melanoma) compared with all other solid tumours, making it difficult to identify cancer driver genes.11 Ultraviolet radiation is the dominant carcinogen in skin carcinogenesis, causing characteristic UVB-induced dipyrimidine dimers in a recognized UV signature,12 as well as UVA-associated oxidative damage coupled with local immunosuppression. Recent genomic studies of cSCC have identified the importance of NOTCH 1 and 2 gene mutations as well as TP53, CDKN2A mutations with a low level of activated H Ras and K Ras mutations.11, 13 Inactivating mutations of TGFBR1 and 2 have also been found.13 Unfortunately, this loss of tumour suppressor gene function is difficult to target therapeutically although there has been a suggestion that oncogene activation occurs in metastatic cSCC.14 Mutations in these key driver genes including also FAT 1 and FGFR3 have been found in apparently normal sun-exposed skin,15 as well as in actinic keratoses,16 which underlies the field cancerization and suggests that sun-exposed skin is primed to form ‘pre-pro cancers.’15 Patients who are immunosuppressed as organ transplant recipients (OTRs) or for treatment of autoimmune disease have long been recognized to be at increased risk of skin cancers with a reversal of the normal ration of BCC to cSCC such that there is an increased risk of 100- to 250-fold in cSCC4, 17, 18 and a high risk of metastasis.19 The role of beta and gamma papillomaviruses has been extensively studied20, 21 and is thought to play an early role in the abrogation of UV-induced apoptosis in allowing the escape and expansion of mutant clones.22 However, there is some epidemiological evidence that certain drugs play a role in the increased risk for OTRs and the recent finding of an azathioprine signature in the DNA of such tumours tends to support this concept.23 As many actinic keratoses do not progress and may indeed regress, and as many cSCCs do not metastasize, there is a need to identify those patients at high risk of progression for potential use of adjuvant therapy and more intense supervision. Conversely, treatment pathways may also be streamlined in patients at low risk of progression to reduce the pressure on elderly patients and an overloaded clinical service. There have been multiple cSCC classifications based on specific histological criteria such as tumour differentiation, depth of invasion, and perineural invasion. It is also well known clinically that certain sites are at high risk such as lip and ear,24 and management protocols are based around the importance of excisional and Moh's surgery.25 The population of patients with cSCC and metastatic cSCC is elderly, often with multiple morbidities and having had multiple primary tumours and so general medical considerations will influence the feasibility of aggressive treatment. OTRs tend to be younger as tumours occur approximately 15 years earlier in this population, but they are particularly predisposed to have multiple primary cSCC. In the era of personalized medicine, it is hoped that new insights to identify patients at risk of progression can lead to streamlined treatment pathways. In this context, it is interesting that SCCs of other body sites have a ‘Commonality of genomic, genetic and epigenetic alterations and similar impact of underlying mesenchyme’26 which supports the hypothesis that there is a common pan-squamous signature.27 Many cancer/SCC studies do not include cSCC, but we can gain insights from SCCs of other organs, particularly relating to the loss of differentiation-associated genes and a need for pro-differentiation agents. Immunotherapies have been particularly important in the treatment of cutaneous melanoma from cancer vaccines to immune cell-related infusions and recent drug development.28 The development of immune checkpoint inhibitors as well as BRAF and MEK inhibitors has provided hope at last for patients with metastatic melanoma.29 There are sound biological reasons for considering that cSCC, with a high tumour mutational burden and increased neoantigen production, would be a good candidate also for immunotherapy and so much attention is focusing on immune mechanisms in cSCC.30 The fact that loss of immune surveillance following immunosuppressive therapy leads to such a dramatic increase in cSCC would further support this argument, although such patients would presently be excluded from such therapy because of the risk of transplant rejection. Ongoing and early clinical trials are extremely encouraging of this approach31 and will inform how we can best treat in future these patients with advanced cSCC, who have seen little therapeutic benefit from chemotherapy. Advanced cSCC is going to increase as the demography of the at-risk population changes with increasing longevity. Although fortunately many patients can be helped by surgical treatment, patients with a high risk of metastasis need to be identified and treatment strategies will further develop with new approaches to immuno- and targeted therapy.