Outcomes of surgery in immunocompromised patients with parotid metastases from cutaneous head and neck squamous cell carcinomas

Introduction

Cutaneous squamous cell carcinoma (cSCC) is the most common malignancy in Australia with 90% occurring on the frequently sun-exposed head and neck region (1). The parotid gland is a site of locoregional spread as intra-parotid lymph nodes receive drainage from the cheek, auricle, temple and forehead regions. Metastatic cutaneous head and neck squamous cell carcinoma (cHNSCC) to the parotid gland indicates more advance stage disease and is associated with poorer prognosis (2). It is the most common malignant tumour found in the parotid gland in Australia (3).

In metastatic cHNSCC to the parotid, standard management is surgery followed by adjuvant radiotherapy. The extent of surgery includes primary tumour resection, parotidectomy and neck dissection. In more advanced tumours, resection may also include the external auditory canal, temporal bone and infratemporal fossa. When the facial nerve is clinically compromised or invasion is seen intraoperatively, it is sacrificed and followed proximally until clear of disease.

Immunosuppressed patients are 65 to 100 times more likely to develop cSCC compared to immunocompetent individuals (4). Immunosuppression is a risk factor in cHNSCC for parotid metastases (5) and is associated with increased recurrence rates and reduced survival (6). Compared to immunocompetent patients, immunosuppressed individuals exhibit a higher incidence of adverse pathological features including lymphovascular and perineural invasion, extracapsular extension (ECE) and poorly differentiated disease (7).

Most existing studies examining this patient cohort of immunosuppressed patients with cHNSCC with metastases to the parotid have been limited by small sample size, ranging from 12 to 33 participants (8-10), resulting in uncertainty regarding the optimal therapeutic strategies for this group of patients. Given their worse prognosis, the benefit of aggressive morbid surgical and adjuvant treatment, remains unclear. Additionally newer therapeutic options such as immunotherapy is seldom offered to those with solid organ transplants given the uncertain evidence of strong benefit balanced with the risk of graft rejection (11), though this is an evolving space with promising results in renal transplants and high risk malignancies (12). This highlights the importance in understanding the relationship between the immune system and cSCC.

The objective of this study is to determine the outcomes of immunocompromised versus immunocompetent patients with metastases to the parotid gland from cHNSCC following parotidectomy. We hope to use differences in outcomes found between the two groups, to inform future treatment decisions for these patients. This single institution (Alfred Hospital, Melbourne, Australia) setting is the state-wide service for sub-specialty units including heart and lung transplant, malignant haematology and human immunodeficiency virus (HIV), and consequently manages a high burden of immunosuppressed patients.

Methods

The study is reported according to the STrengthening of Reporting of OBservational studies in Epidemiology (STROBE) reporting guidelines (available at https://www.theajo.com/article/view/10.21037/ajo-2025-1-70/rc). The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Ethics approval was obtained from the Alfred Health Ethics Committee (249/24) for a retrospective longitudinal cohort study. Informed consent was waived because this was deemed a low-risk, retrospective analysis of previously collected clinical information, for which obtaining individual consent was impractical. A consecutive series of patients that underwent parotidectomy between February 2008 and June 2021 was obtained from Cerner SurgiNet electronic database with Medicare Benefits Schedule (MBS) codes 30250, 30253, 30251 and 30247 (Appendix 1).

Patients were included if they had histopathologically confirmed regional metastatic cHNSCC involving the parotid gland and had undergone parotidectomy with a minimum of 2 years of clinical follow-up. To ensure case completeness, these records were cross-referenced with institutional head and neck multidisciplinary meeting agendas. Patients were excluded if histopathological analysis revealed a benign lesion or a non-cutaneous primary tumour.

Patients were stratified based on immune status, as determined from comprehensive review of electronic medical records. For the purposes of this manuscript, any patient with solid organ transplantation, haematological malignancy, HIV infection, splenectomy, severe chronic kidney disease (CKD) (13), long-term immunosuppressive therapy and diabetes with evidence of poor perioperative glycaemic control (Table 1), were assumed to be immunosuppressed. Those without these conditions were determined to be immunocompetent.

Demographic, pathological, and follow-up data were extracted from electronic medical records. Staging was conducted using the O’Brien P staging system, which was selected for its greater specificity in characterising parotid nodal disease compared to the conventional tumor-node-metastasis (TNM) system (Appendix 2) (3).

Statistical analysis

The Wilcoxon-Mann-Whitney test was used to compare continuous outcomes between two groups. For dichotomous categorical variables, Fisher’s exact test was performed if a table had 25% of cell counts less than five. Otherwise, the Chi-squared (χ²) test was utilized. Univariable survival analyses were conducted using the log-rank test.

Survival analysis ensured that all patients contributed meaningful data from initial parotidectomy to the end of follow-up. Purposeful variable selection method was used to incorporate variables with a P value <0.2 from univariable analyses into multivariable analyses. Given that patients were not randomised, we used multivariable analyses wherever possible to adjust for baseline differences between the two cohorts. The included logistic regression for binary outcomes, reported as odds ratios (ORs) and Cox-Hazards regression for overall survival (OS), reported as hazard ratios (HRs). For multivariableanalysis of locoregional recurrence (LRR), death and distant metastases were treated as competing events, as they could occur simultaneously. Likewise, analysis of distant metastases also treated death as a competing event. Competing-risks regression was conducted using the Fine-Gray model, reported as subdistribution hazard ratios (sHR) and illustrated using cumulative incidence functions (CIF). A significance level of P value <0.05 was used. All statistical tests were performed using Stata version 19.0 (StataCorp, College Station, Texas).

Results

Ninety-five patients with cHNSCC were identified with confirmed parotid metastases, 35 were immunocompromised (Table 1) and 60 were immunocompetent. All patients (Table 2) included in this study underwent parotidectomy with 78.9% (75/95) limited to the superficial lobe. Additional procedures included neck dissection (71.6%, 68/95), mastoidectomy (10.5%, 10/95) and lateral temporal bone resections (9.5%, 9/95). Post-operative radiotherapy was administered to 92.6% (88/95) and additional adjuvant chemotherapy to 6.3% (6/95). Compared to immunocompetent patients, the immunocompromised were more likely to be male (94.3% vs. 73.3%; P=0.012) and younger at the time of surgery (median age 75 vs. 77 years; P=0.028). They had lower stage (P1) parotid tumours (41.7% vs. 62.9%; P=0.028) at time of surgery. However the median size of the metastatic tumour deposit in the parotid was not different between the two groups, 24 mm in immunocompromised patients versus 25 mm in immunocompetent patients (P=0.7). Immunocompromised patients were more likely to have lymphovascular invasion on histopathology (53.1% vs. 30.2%; P=0.045) (Table 2).

The median time to LRR for the entire cohort was 31 months (IQR, 9–55 months) (Table 3). This was significantly shorter in the immunocompromised group at 15 months (IQR, 5–39 months), compared with 47 months (IQR, 17–67 months) in the immunocompetent cohort (P<0.001). The median time to death was 16 months (IQR, 7–30 months) in immunocompromised patients and 20 months (IQR, 14–39 months) in immunocompetent patients, with no significant difference between groups (P=0.607). The median follow-up duration for the overall cohort was 45 months (IQR, 20–64 months), and all patients were included in the subsequent analyses.

Facial nerve sacrifice

Facial nerve sacrifice was required in 17.9% (17/95) of patients (Table 2). Facial nerve sacrifice did not impact survival or recurrence (Table 4).

OS

The 2-year survival rates were 60.0% for immunocompromised and 76.7% for immunocompetent. In the multivariable analyses adjusting for patient age, tumour stage, disease recurrence and adjuvant therapy, risk of death in immunocompromised patients was not statistically different to immunocompetent, but there was a trend towards increased risk [HR 2.10; 95% confidence interval (CI): 0.86–5.12; P=0.104]. Over a 5-year period however (Figure 1), there was twice the risk of death (HR 2.06; 95% CI: 1.02–4.19; P=0.045) in immunocompromised when compared to immunocompetent patients. The 5-year survival rates were 29.4% for immunocompromised and 62.2% for immunocompetent. Other factors that led to death in the immunocompetent cohort include pneumonia, ischaemic heart disease and cerebrovascular disease. This cohort was also generally older compared to immunocompromised patients (median age 77 vs. 75 years; P=0.028).

Figure 1 Kaplan-Meier estimates of overall survival following parotidectomy for cutaneous head and neck squamous cell carcinoma metastatic to the parotid gland, stratified by immunocompromised versus immunocompetent status. Vertical marks: number of censored observations. IC, immunocompromise.

Disease-specific survival (DSS) and disease-specific mortality (DSM)

The 2-year DSS was 76.2% in immunocompromised and 85.9% in immunocompetent. The 5-year DSS was 59.9% in immunocompromised and 75.38% in immunocompetent. After adjusting for potential confounders in multivariable analysis, immunocompromised status was not statistically significantly associated with higher DSM over a 2-year (sHR 1.62; 95% CI: 0.42–6.26; P=0.49) and 5-year follow-up period (sHR 1.91; 95% CI: 0.64–5.75; P=0.248).

LRR

In multivariable analysis adjusting for patient age, tumour stage and adjuvant therapy , immunocompromised status was independently associated with a significantly higher cumulative incidence of LRR at both 2 years (sHR 3.27, 95% CI: 1.49–7.19, P=0.003) and 5 years (sHR 2.75, 95% CI: 1.32–5.76, P=0.007). At 5 years, the cumulative incidence of LRR was approximately 43% in immunocompromised patients, compared to 18% in immunocompetent patients. This is shown in Figure 2.

Figure 2 Cumulative incidence of locoregional recurrence following parotidectomy for cutaneous head and neck squamous cell carcinoma metastatic to the parotid gland, stratified by immunocompromised versus immunocompetent status.

Distant metastasis (DM)

In multivariable analyses adjusting for patient age, tumour stage and adjuvant therapy, immunocompromised status was not associated with increased cumulative incidence of DM at 2 years (sHR 0.77, 95% CI: 0.20–2.96; P=0.70) and 5 years (sHR 0.74, 95% CI: 0.21–2.57; P=0.64). Nodal stage N2 was independently associated with a higher incidence of DM (sHR 3.29, 95% CI: 1.21–8.93; P=0.02). At 5 years, the incidence of metastases, accounting for deaths, were approximately 3% for immunocompromised patients and 4% for immunocompetent patients (Figure 3).

Figure 3 Cumulative incidence of distant metastasis following parotidectomy for cutaneous head and neck squamous cell carcinoma metastatic to the parotid gland, stratified by immunocompromised versus immunocompetent status.

Discussion

This study reports on patients with cHNSCC with parotid metastases that required surgical intervention with a parotidectomy. Being immunocompromised was independently associated with significantly lower OS and higher LRR. The numbers of immunocompromised patients in similar studies seen in Table 5, are low ranging from 12–33 patients (8–25%) (8-10). In our study, numbers are similarly low but the largest published, with 35 immunocompromised patients with cHNSCC with parotid metastases, which was 36.8% of our cohort studied. The proportion of immunocompromised patients is higher in our study (Table 5), with one in three patients analysed, we assume due to our institutions high burden of immunocompromised patients.

Immunocompromised patients were more likely to have lymphovascular invasion on histopathology (P=0.045), which is associated with poorer outcome. Notably, there was no significant difference in ECE, perineural invasion and positive margins in the two groups. We would expect these factors to be associated with a poorer outcome, but this could be due to limited power.

Our multivariable analysis demonstrated a statistically significantly poorer OS in immunocompromised patients, with a rate of 34.3% and twice the risk of death at 5 years (HR 2.06; 95% CI: 1.02–4.19; P=0.045). Our 5-year OS rates are similar to, but higher than other similar studies (8,9), although this may be due to a higher proportion of chronic lymphocytic leukaemia (CLL) patients in their cohorts. CLL patients are known to have more aggressive SCC and poorer prognosis (14). The true OS for all patients with metastatic SCC at our centre is likely to be poorer as only patients that underwent surgery were included in our study, therefore not capturing patients with unresectable disease or those too comorbid for surgery. Immunocompromised patients demonstrated a significantly increased risk of LRR, with a significantly higher hazard at both 2 years (sHR 3.27) and 5 years (sHR 2.75) compared to immunocompetent patients. The cumulative LRR rate at 5 years was 43% in the immunocompromised group. Our results for 5-year DSS and LRC are similar to other studies.

Several large studies of cHNSCC have analysed the effect of immunosuppression on outcomes. Zavdy et al. studied 465 patients (335 immunosuppressed) and Tam et al. (15) studied 796 patients (147 immunosuppressed), both showing immunocompromised patients had decreased OS and higher rates of recurrence, consistent with the present study. However, neither study separated out parotid metastases specifically and may have had a larger proportion of non-metastatic cHNSCCs studied. This restricts direct comparability to our study. Additionally, they had a broader inclusion criterion for immunocompromise; Zavdy included systemic lupus erythematosus and psoriasis patients (14), whilst Tam included those with type 1 or 2 diabetes mellitus treated with insulin in their immunocompromised groups (15). A recent study by Wuthikraikun et al. (16) found that cancer patients with diabetes mellitus and poor glycaemic control have worse survival outcomes compared to the non-diabetic and even diabetic patients with good glycaemic control. There were five patients with diabetes involved in the study. Three of them also had another reason for immunocompromise. The remaining two had insufficient data regarding peri-operative glycaemic control, a limitation of the retrospective design, and therefore were included in the immunocompetent cohort.

Immunocompromised patients were more likely to present with P1 disease (P=0.028), defined as a metastatic parotid deposit less than 30mm. However, when the exact size of the parotid tumour deposit was analysed, there was no significant difference between the median size in both of the groups. A study by Veness et al. (17) also saw median primary tumour size was reduced in immunocompromised patients (12mm) compared to immunocompetent patients (15 mm). Also, Zavdy et al. found that those with solid organ transplants were more likely to have smaller primary tumour deposits (median 13.5 mm) compared to other immunosuppressed and immunocompetent groups (14). They postulated this may be due to regular skin checks in a patient cohort already well linked with medical services and good screening protocols in this high-risk group (14). Similar reasoning would apply in our patient cohort.

Immunocompromised patients were more likely to receive chemotherapy (P=0.015), however the small sample size precludes meaningful statistical analysis. Our institution typically avoids the use of post-operative concurrent chemoradiotherapy given the limited evidence of benefit (18). Immunotherapy, is judiciously offered to immunocompromised patients, especially those with solid organ transplants, given the uncertain evidence of strong benefit balanced with the risk of graft rejection (11). Therefore, chemotherapy is a palliative treatment option in these patients (19). Immunotherapy is an expanding systemic therapy to manage cHNSCC, hence the need to better understand the effects of immunosuppression on the disease.

At our institution, we follow up our head and neck cancer cases every three months in the first 2 years. Then between 4 to 6 monthly from years 3 to 5. More frequent follow-up for high-risk immunosuppressed patients should be considered such as second monthly in the first 2 years, and three monthly reviews until the fifth year. Our results demonstrate a three-fold increased risk of recurrence as well as a propensity for early recurrence in the immunocompromised group, with a median time to recurrence of 15 months compared to 46.5 months in the immunocompetent group (P≤0.001).

Facial nerve sacrifice was required in 17.9% (17/95), with no impact on survival or recurrence in either group analysed. This is consistent with the results of Shao et al. (9) who studied 35 cases of metastatic cHNSCC that required facial nerve sacrifice and found no impact on disease free survival on multivariate analysis. Given the poor prognosis in immunocompromised patients, avoiding aggressive morbid surgery should be considered in this group if presenting with bulky metastatic disease and facial nerve involvement. These decisions will involve multidisciplinary discussion and consideration of other patient and disease factors. A quality of life study would be of value in this group of patients.

This study had several limitations, including its retrospective design. This risks incomplete data entry, such as the lack of information regarding peri-operative glycaemic control. Loss to follow-up is an issue once patients are referred back to their local peripheral centres. Thankfully, our institute has a strong relationship with the peripheral radiation oncology department, with regular correspondence between the centres. Information less relevant to the disease such other causes of death may have been incompletely gathered. Also, our results are limited by relatively small sample size due to single centre data, albeit a large tertiary centre with a higher proportion of immunocompromised patients. This reduces the generalisability of the study to the rest of the world. This was mitigated with the long period of recruitment and without a heavily restrictive inclusion criterion. As seen in Table 5, similar studies suffer difficulties with capturing immunocompromised patients with cHNSCC treated with parotidectomy. Future multicentre studies may help clarify the significance of this further.

Conclusions

Immune status is an important factor in patients who underwent surgery for cHNSCC with metastases to the parotid gland. A compromised immune system leads to more aggressive tumour biology, resulting in poor prognosis. More frequent follow up for immunocompromised patients is recommended as disease recurrence was more common and developed earlier in immunocompromised patients. OS is poor in immunocompromised patients with standard surgical and radiation treatment and therefore cancer institutions seeing this patient population regularly should carefully consider the use of aggressive curative treatment where a reasonable chance of cure may be difficult to achieve.

Acknowledgments

The authors would like to acknowledge Min Ting Tan for contribution to data collection.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://www.theajo.com/article/view/10.21037/ajo-2025-1-70/rc

Data Sharing Statement: Available at https://www.theajo.com/article/view/10.21037/ajo-2025-1-70/dss

Peer Review File: Available at https://www.theajo.com/article/view/10.21037/ajo-2025-1-70/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://www.theajo.com/article/view/10.21037/ajo-2025-1-70/coif). J.D.P. serves as an unpaid editorial board member of Australian Journal of Otolaryngology from August 2024 to December 2026. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Ethics approval was obtained from the Alfred Health Ethics Committee (249/24) for a retrospective longitudinal cohort study. Informed consent was waived because this was deemed a low-risk, retrospective analysis of previously collected clinical information, for which obtaining individual consent was impractical.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Badlani J, Gupta R, Smith J, et al. Metastases to the parotid gland - A review of the clinicopathological evolution, molecular mechanisms and management. Surg Oncol 2018;27:44-53. [Crossref] [PubMed]
2. Ch'ng S, Maitra A, Allison RS, et al. Parotid and cervical nodal status predict prognosis for patients with head and neck metastatic cutaneous squamous cell carcinoma. J Surg Oncol 2008;98:101-5. [Crossref] [PubMed]
3. O'Brien CJ, McNeil EB, McMahon JD, et al. Significance of clinical stage, extent of surgery, and pathologic findings in metastatic cutaneous squamous carcinoma of the parotid gland. Head Neck 2002;24:417-22. [Crossref] [PubMed]
4. Pritchett EN, Doyle A, Shaver CM, et al. Nonmelanoma Skin Cancer in Nonwhite Organ Transplant Recipients. JAMA Dermatol 2016;152:1348-53. [Crossref] [PubMed]
5. O'Hara J, Ferlito A, Takes RP, et al. Cutaneous squamous cell carcinoma of the head and neck metastasizing to the parotid gland--a review of current recommendations. Head Neck 2011;33:1789-95. [Crossref] [PubMed]
6. Southwell KE, Chaplin JM, Eisenberg RL, et al. Effect of immunocompromise on metastatic cutaneous squamous cell carcinoma in the parotid and neck. Head Neck 2006;28:244-8. [Crossref] [PubMed]
7. Manyam BV, Garsa AA, Chin RI, et al. A multi-institutional comparison of outcomes of immunosuppressed and immunocompetent patients treated with surgery and radiation therapy for cutaneous squamous cell carcinoma of the head and neck. Cancer 2017;123:2054-60. [Crossref] [PubMed]
8. McDowell LJ, Tan TJ, Bressel M, et al. Outcomes of cutaneous squamous cell carcinoma of the head and neck with parotid metastases. J Med Imaging Radiat Oncol 2016;60:668-76. [Crossref] [PubMed]
9. Shao A, Wong DK, McIvor NP, et al. Parotid metastatic disease from cutaneous squamous cell carcinoma: prognostic role of facial nerve sacrifice, lateral temporal bone resection, immune status and P-stage. Head Neck 2014;36:545-50. [Crossref] [PubMed]
10. Schmidt C, Martin JM, Khoo E, et al. Outcomes of nodal metastatic cutaneous squamous cell carcinoma of the head and neck treated in a regional center. Head Neck 2015;37:1808-15. [Crossref] [PubMed]
11. Trinconi Cunha M, Wallace N, Porceddu S, et al. Top advances of the year: Developments of immunotherapy in cutaneous squamous cell carcinoma, 2023-2024. Cancer 2025;131:e35920. [Crossref] [PubMed]
12. Legris T, Sallée M, Charmetant X, et al. Immune Checkpoint Inhibitors in Kidney Transplant Recipients: A French Multicenter Retrospective Cohort Study. Transplant Direct 2025;11:e1851. [Crossref] [PubMed]
13. Syed-Ahmed M, Narayanan M. Immune Dysfunction and Risk of Infection in Chronic Kidney Disease. Adv Chronic Kidney Dis 2019;26:8-15. [Crossref] [PubMed]
14. Zavdy O, Coreanu T, Bar-On DY, et al. Cutaneous Squamous Cell Carcinoma in Immunocompromised Patients-A Comparison between Different Immunomodulating Conditions. Cancers (Basel) 2023;15:1764. [Crossref] [PubMed]
15. Tam S, Yao CMKL, Amit M, et al. Association of Immunosuppression With Outcomes of Patients With Cutaneous Squamous Cell Carcinoma of the Head and Neck. JAMA Otolaryngol Head Neck Surg 2020;146:128-35. [Crossref] [PubMed]
16. Wuthikraikun C, Panja P, Decha-Umphai C, et al. Impact of diabetes and glycemic control during cancer care on mortality in non-metastatic solid and hematologic malignancies. Ann Med 2025;57:2575108. [Crossref] [PubMed]
17. Veness MJ, Palme CE, Morgan GJ. High-risk cutaneous squamous cell carcinoma of the head and neck: results from 266 treated patients with metastatic lymph node disease. Cancer 2006;106:2389-96. [Crossref] [PubMed]
18. Porceddu SV, Bressel M, Poulsen MG, et al. Postoperative Concurrent Chemoradiotherapy Versus Postoperative Radiotherapy in High-Risk Cutaneous Squamous Cell Carcinoma of the Head and Neck: The Randomized Phase III TROG 05.01 Trial. J Clin Oncol 2018;36:1275-83. [Crossref] [PubMed]
19. Martinez JC, Otley CC, Okuno SH, et al. Chemotherapy in the management of advanced cutaneous squamous cell carcinoma in organ transplant recipients: theoretical and practical considerations. Dermatol Surg 2004;30:679-86. [Crossref] [PubMed]