Squamous Cell Carcinoma of Scalp: Our Experience in a Single Tertiary Care Centre

Abstract

Cutaneous cancer is a rare entity accounting for less than 1% of all diagnosed cancers in India. In contrast to the West, squamous cell carcinoma is the commonest skin cancer in India, often affecting the scalp. However, due to their rarity, not much is known regarding their biological behavior and prognosis. The present study is a retrospective cohort study undertaken in a tertiary cancer centre on 19 consecutive cases of squamous cell carcinoma of the scalp over a period of 5 years. Patients with non-sqamous histology and those associated with xeroderma pigmentosum were excluded. Nineteen patients were evaluated with a mean age of 52.7 years. Majority presented with a swelling (11 patients; 57.9%) in the parieto-occipital region (13 patients; 68.4%). All patients underwent wide excision with or without excision of underlying bone or dura, depending on involvement. Only 4 (21%) required major reconstruction. On histopathology, 8 (42.1%) were poorly differentiated, while 3 (15.8%) had presence of perineural invasion (PNI). The mean duration of follow-up was 38.14 months. Three of the 4 (75%) and 2 of 3 (66.67%) patients with poorly differentiated histology and PNI respectively developed recurrence, while only 1 of the 6 (16.67%) with close or positive base margin developed recurrence. Hence, to conclude, poorly differentiated histology and PNI are poor prognostic indicators, while in the event the mucoperiosteum is clinically uninvolved, the underlying bone may be preserved in select cases, even if the base margin is close.

Introduction

Cancer of the skin is a rare entity and comprises less than 1% of all diagnosed cancers in India. Over the last few decades, however, the incidence has been rising steadily. Basal cell carcinoma (BCC) and Squamous cell carcinoma (SCC), together known as Non-Melanomataous Skin Cancer (NMSC), and malignant melanoma are the most common cutaneous cancers, of which SCC is the commonest skin cancer in India. This is in contrast to the West, where BCC is the commonest cutaneous malignancy [1, 2]. In the head and neck, cutaneous SCC most commonly involves the ear, frontotemporal region, and cheek [3]. However, much is not known regarding its behavior and prognosis.

The present study was undertaken in an attempt to evaluate the pathology of cutaneous SCC involving the scalp, and its response to treatment.

Materials and Methods

A retrospective cohort study was conducted of individuals diagnosed with squamous cell carcinoma of the scalp, which presented to a tertiary care hospital over a period of 5 years from 2005 to 2009. Patients with non-squamous histology and those associated with xeroderma pigmentosum were excluded from the study.

The demographic and pathological data collected from the medical records included patient age, gender, history of addiction, family history, previous treatment received, and duration of symptoms. Note was made of the location of the lesion and the presence or absence of lymph node metastases. Locations were designated as frontal, temporal, parietal, or occipital regions of the scalp. The treatment received and pathological report of the patient was also documented, while the follow-up status was recorded from the medical records and telephonic contact with the patient or relatives at the time of final assessment.

Results

Demographic Details

A total of 19 patients were included in the present study, of which 10 (52.6%) were males, and 9 (47.4%) were females, with a mean age of 52.7 years (ranging from 34 to 85 years). Seven patients (36.8%) had history of addictions, of which 6 had history of addiction to tobacco, while one patient had history of tobacco and alcohol addiction. Only 3 patients (15.8%) had family history of malignancy in first degree relatives. The mean duration of symptoms was 23.7 months, ranging from 2–108 months.

The presenting complaint was a swelling in 11 (57.9%), ulcer in 5 (26.3%), and ulceroproliferative lesion in 3 (15.8%) patients. Most of the lesions involved the parietal and occipital regions of the scalp, in 6 (31.6%) and 7 (36.8%) patients respectively, together comprising 68.4% of total lesions. The individual site-wise distribution of lesions is given in Table 1. Overall only 3 (15.8%) patients had palpable lymph node metastases, all involving level V, while the remaining 16 patients were clinico-radiologically node negative.

Table 1.

Site-wise distribution of scalp lesions

SITE	n (%)
Parietal	6 (31.6)
Occipital	7 (36.8)
Frontal	2 (10.5)
Temporal	1 (5.3)
Vertex	3 (15.8)
Total	19

Ten patients (52.6%) had history of receiving prior treatment, of which 9 had received some surgical intervention, while one had received radiotherapy. Among the 9 patients previously operated, 7 had undergone excision of the lesion, while 2 had undergone multiple excisions, including removal of the underlying bone.

Surgery Details

All cases underwent wide excision with 1 cm margins. The underlying bone was also excised in 9 (47.4%) due to clinical evidence of infiltration by the disease, of which only the outer table was excised in 3 patients, while full thickness was removed in the remaining 6 patients. Of these 6 patients, the underlying dura was also resected in 2 patients. Lymph node dissection was done in 3 patients, of which two patients underwent posterolateral neck dissection and occipital node clearance, while the remaining patient had occipital node sampling, which was negative for metastasis on frozen section.

Histopathological Details

1. Differentiation: Majority of the lesions i.e. 8 patients (42.1%) were poorly differentiated, followed by 7 (36%) patients moderately differentiated, and 4 (21%) were well differentiated. Three (15.8%) patients had evidence of perineural invasion, while none had evidence of lymphovascular emboli on histopathology.

2. Margin status: The lateral cut margin was positive in only 1 patient (5.3%), while the rest were free, i.e. more than 5 mm. On examining the base, 3 (15.8%) were positive. All had full thickness bone removed during surgery, and one even had dura removed due to clinical evidence of infiltration. Three patients (15.8%) had close margin of base. This was usually the case when the entire thickness of the skin was excised with underlying mucoperiosteum, but the underlying bone was clinically not infiltrated, and was hence not included in the resection.

3. Reconstruction: The surgical defect in 13 (68.4%) patients was closed by local rotation, 4 (21%) required microvascular free tissue transfer, while 2 (10.5%) received split thickness skin graft.

Adjuvant Treatment and Follow-Up

The mean follow-up duration was 38.14 months (range 1–141 months). Four patients were advised adjuvant radiotherapy. This included all patients with positive margins, three of which also had perineural invasion, and one additional patient with recurrent tumour and presence of perineural invasion. One of these patients refused postoperative radiotherapy, and went on to develop locoregional recurrence and lung metastases 2 months later, and received palliative chemotherapy for the same. Another patient with similar disease, but who received post operative radiotherapy, also developed local recurrence with bone metastasis 6 months later. Both these patients succumbed to the disease.

Of the remaining patients, two developed local recurrences two and three months later respectively, and were subjected to wide excision with 1 cm margins. One of these again developed local recurrence 6 months later, which was treated with wide local excision. Both these patients were disease-free at the time of final evaluation, even 10 years after the completion of treatment. Of the other patients, five were lost to follow-up after treatment, while one died of other causes. Among the remaining patients, 4 were disease-free even 10 years after the initial treatment. However, the remaining 5 patients were followed-up only 5 years after the initial treatment. None of these 5 patients had any evidence of locoregional or distant recurrence at the time of last follow-up.

Hence, it can be seen that 3 of the 4 patients (75%) and 2 of the 3 (66.67%) patients with poorly differentiated carcinoma and perineural invasion respectively developed recurrences on follow-up, while only one of the 6 (16.67%) patients with either close or positive base margin developed recurrence on follow-up.

Discussion

Non melanoma skin cancer (NMSC) is one of the commonest cancers in the West, with an incidence of 1 million new cases per year in the United States [4]. In India, however, cutaneous malignancy accounts for less than 1% of all cancers. The incidence has been increasing in the recent years, with some authors reporting an incidence of up to 3.18% in Northern India [1, 5]. The scalp is involved by approximately 2% of all skin cancers [6]. Dark skinned individuals are more prone to develop SCC rather than BCC, which may explain why SCC is the commonest cutaneous malignancy in India as compared to the rest of the world, where BCC is the commonest histology of skin cancer.

Cutaneous SCC is thought to mainly affect males, since it is believed that women are protected from sun exposure by relatively longer hair and reduced incidence of alopecia [7]. However, few recent studies report a higher incidence in females [5]. In our study too, there was a nearly equal gender distribution, with 10 males and 9 females in the study population.

Most of the lesions in the present study involved the parieto-occipital region of the scalp, comprising of 68.4% of total lesions. This is in accordance with observations of Denewer et al. [8] who also had majority of lesions in the lateral regions of the scalp, while Katz et al [7] and Sweeny et al. [9] had the vertex and frontal scalp as major sites of involvement.

Following excision, only 4 patient s required major reconstruction in the form of microvascular free tissue transfer, while in the remaining 15 patients, local rotation flap or split thickness skin grafting was sufficient. This was also the case in patients who presented to us with past history of excisions, or those who underwent excision for local recurrences during follow-up.

In our series, most of the tumours (42.1%) were poorly differentiated, which is considered as one of the high risk features in the AJCC staging for cutaneous malignancy. Three of the four patients which developed recurrences on follow-up had poorly differentiated tumours. In addition, perineural invasion, also considered as a poor prognostic factor, was also seen in three patients, 2 of which developed recurrences of follow-up, thus confirming the adverse influence of both these tumour factors on prognosis.

On the contrary, margin status did not have a very significant effect on recurrence, especially for the base margin which was found to be either close or positive in 6 patients (31.6%).This was usually the case when either the whole, or a significant thickness of the scalp skin was involved by the disease while the mucoperiosteum and underlying bone were clinically uninvolved, and hence were not included in the resection. Of these cases, only one patient, who refused adjuvant radiotherapy, went on to develop recurrence.

Conclusion

Squamous cell carcinoma of the scalp is a relatively uncommon disease, affecting both males and females equally. The primary treatment is with wide local excision of the lesion, with neck dissection when indicated. Poorly differentiated histology and perineural invasion are associated with aggressive tumour behavior, responsible for multiple local recurrences. However, unlike other head neck malignancies, these are often amenable to local excision, often without the need for major reconstruction and without affecting overall prognosis. Local rotation flaps and split thickness skin grafting are usually sufficient for reconstruction, with microvascular free tissue transfer reserved for larger defects. Although achieving negative margins during excision is paramount, in the event the mucoperiosteum is clinically uninvolved, the underlying bone may be preserved in select cases even if the base margin is close.

The present study has a few limitations including the retrospective nature of the study. Also, the study population comprises of a small number of patients treated over a long period of 5 years. Although the authors agree that a retrospective study is not ideal and the number of patients is small, but this is an uncommon disease in India. If a tertiary cancer centre only sees 19 cases in 5 years there is little chance of a large prospective study being performed.

Compliance with Ethical Standards

Conflict of interest

All author declares that they have no conflict of interest.