Secondary Logo

Journal Logo

Original Article

Keratinocyte Carcinoma Mortality in the United States as Reported in Death Certificates, 2011–2017

Misitzis, Angelica MD*,†; Beatson, Meghan BS*,†; Weinstock, Martin A. MD, PhD*,†

Author Information
doi: 10.1097/DSS.0000000000002261
  • Free

Abstract

Keratinocyte carcinoma (KC) consists of basal cell carcinoma (BCC) and squamous cell carcinoma (SCC). Keratinocyte carcinoma has a considerable burden in the United States; it is the most common malignancy in white skin populations, and its incidence has been rising.1 It was estimated in 2012 that 3.3 million individuals were diagnosed with 5.4 million KCs in the United States.2 Furthermore, considerable disfigurement and morbidity can occur before they are treated. Previous reports showed declining KC mortality rates (MRs) until 2000, followed by a leveling; the age-adjusted US MR was at 1.1 1980, dropped to 0.5 in 2000.3,4 The objective of this study is to assess the trends of KC MRs in the United States for years 2011 through 2017.

Methods

Data Collection

The data from the records of the Center of Disease Control and Prevention (CDC) were obtained from WONDER (Wide-ranging Online Data for Epidemiological Research) system for years 2011 to 2017.5 US 2000 Standard population was used. Races included were “Whites,” “Black or African American,” “American Indians or Alaska Natives” and “Asians or Pacific Islanders.” Mortality rates were obtained directly from WONDER, which adjusts them to the US 2,000 population.

To capture deaths attributed to KC, the following ICD-10 codes were included: C44.0 (Skin of lip); C44.1 (Skin of eyelid, including canthus); C44.2 (Skin of ear and external auricular canal); C44.3 (Skin of other and unspecified parts of face); C44.5 (Skin of trunk); C44.6 (Skin of upper limb, including shoulder); C44.7 (Skin of lower limb, including hip); C44.8 (Overlapping lesion of skin); C44.9 (Malignant neoplasm of skin, unspecified); C21.0 (neoplasms of anus); C51 (vulvar cancer); C60 (penile cancer), and C63.2 (scrotum cancer). WONDER allows only for 3-digit searches, so we were not able to separate between BCC and SCC, in each anatomic location.

We excluded C44.4 (neoplasms of the scalp and neck), because a previous analysis of death certificates showed that 85% of them were SCC of mucosal sites.6 Data are presented in 3 categories as “genital,” (for vulvar, penile and scrotum cancer), “perianal” (for neoplasms of the anus), and KC in chronically or intermittently “sun-exposed” anatomic sites. We followed the procedure of Karia and colleagues7 and Wu and Weinstock,4 States were divided into 2 zones. The northern (lower UV) zone included the following states: CT, ID, IL, IN, IA, ME, MD, MA, DE, MI, MN, NH, NJ, NY, ND, OH, OR, PA, MT, RI, DC, VT, WA, WV, WY SD, AK, and WI. The southern (high UV) zone consisted of the following states: AL, AZ, AR, CA, CO, FL, GA, KS, KY, LA, MS, MO, NE, NV, NM, NC, OK, SC, TN, TX, UT, VA, and HI.

Statistical Analyses

Univariate linear regression analyses (dependent variable: number of deaths, independent variable: years) were conducted with Stata 14.1 Software (StataCorp, College Station, TX). Statistically significant p-values were < .05.

Results

In the United States for years 2011 to 2017, KC MR was 1.24 per 100,000 persons per year. Mortality rates were increasing overall (Figure 1) (Liner regression coefficient (β) = 0.038, confidence interval [CI]: 0.028–0.047, p < .001). For the period studied, 31,904 deaths were attributed to KC. In 2017, 2,618 deaths were caused by KC in sun-exposed areas, 1,629 were caused by genital KC and 966 deaths were caused by perianal KC (Table 1).

Figure 1.
Figure 1.:
Age-adjusted keratinocyte carcinoma mortality rates in the United States by anatomic site, 2011 to 2017. For sun-exposed sites, KC MRs increased by 0.1 deaths per 100,000 persons per year over this period. For genital and perianal sites, KC MRs increased by 0.07 deaths per 100,000 persons per year over this period. Rates are age-adjusted to the standard US 2000 population.
TABLE 1. - Keratinocyte carcinoma Age-Adjusted Mortality Rates per 10,000 per Year (Number of Deaths) in the United States, 2011–2017
Sun-exposed Genital Perianal
Males 1.06 (11,359) 0.19 (2,174) 0.19 (2,235)
 Whites 1.19 (10,935) 0.19 (1,857) 0.19 (1,866)
 Blacks 0.25 (304) 0.23 (251) 0.26 (328)
Females 0.31 (4,864) 0.53 (7,754) 0.25 (3,518)
 Whites 0.34 (4,587) 0.58 (7,158) 0.26 (3,153)
 Blacks 0.12 (195) 0.31 (478) 0.20 (306)
Total 0.62 (16,223) 0.38 (9,928) 0.23 (5,753)

Keratinocyte Carcinoma in Sun-Exposed Areas

The age-adjusted MRs for KC in sun-exposed areas for years 2011 to 2017 was 0.62 (men: 1.06, women: 0.31). Overall, KC MR increased by 18% (β: 0.020, CI: 0.008–0.033, p: .008) (Table 2); MRs showed increase in whites (β = 0.023, CI: 0.010–0.035, p: .006), but not in blacks (β = 0.006, CI: −0.006 to 0.019, p: .239). Throughout the period studied, there has been a significant increase in MRs in southern states (β = 0.037, CI: 0.024–0.050, p < .001), but not in the northern states (β = 0.014, CI: −0.008 to 0.037, p: .200). Keratinocyte carcinoma MR for southern states was 0.68 and 0.57 for northern states for the period studied. Crude MRs were higher in older age groups with 0.14 per 100,000 persons per year for patients younger than 65 years, and for patients older than 85 crude MR was 13.14 per 100,000 persons per year (Table 3).

TABLE 2. - Keratinocyte carcinoma Age-Adjusted Mortality Rates per 10,000 per Year (Number of Deaths) in the United States
Year Overall, by KC Type Sun-exposed, by Gender
Genital and Perianal Sun-exposed White Females White Males
2011 0.57 (1,992) 0.56 (1,936) 0.29 (540) 1.08 (1,318)
2012 0.59 (2,066) 0.57 (2,021) 0.33 (609) 1.08 (1,328)
2013 0.56 (2,039) 0.60 (2,106) 0.29 (554) 1.15 (1,472)
2014 0.58 (2,199) 0.64 (2,389) 0.34 (733) 1.24 (1,631)
2015 0.61 (2,289) 0.69 (2,608) 0.38 (696) 1.30 (1,755)
2016 0.66 (2,501) 0.66 (2,545) 0.35 (696) 1.25 (1,729)
2017 0.64 (2,595) 0.66 (2,618) 0.41 (802) 1.20 (1,702)
Regression coefficient, (CI 95%), p-value 0.014, (0.003–0.025), 0.021 0.020, (0.008–0.033), 0.008 0.018, (0.005–0.030), 0.014 0.030, (0.001–0.060), 0.045
KC, keratinocyte carcinoma.

TABLE 3. - Keratinocyte carcinoma Crude Mortality Rates per 10,000 per Year (Number of Deaths) in the United States, 2011–2017
Age Sun-exposed Genital Perianal
<65 0.14 (2,759) 0.13 (2,486) 0.14 (2,764)
65–84 2.78 (7,808) 1.69 (4,745) 0.85 (2,376)
>85 13.16 (5,656) 6.28 (2,697) 1.42 (612)

Genital Keratinocyte Carcinoma

The genital KC MR was 0.38 per 100,000 persons per year for the period studied (males: 0.19, females: 0.53). Overall, KC MR increased by 16% (β = 0.009, CI: 0.004–0.014, p: .005). There was an increase in whites (β = 0.009, CI: 0.001–0.016, p: .027), but no change in blacks (β = −0.0075, CI: −0.020 to 0.005, p: .179). For whites, MRs were borderline increasing in the southern states (β = 0.010, CI: −0.005 to –0.015, p: .064), but not in the northern states (β = 0.005, CI: −0.005 to –0.015, p: .305). Keratinocyte carcinoma MR for southern states was 0.35 and 0.43 for northern states for the period studied. Crude MR was 0.13 per 100,000 persons per year for patients younger than 65 years and for patients older than 85 crude MR was 6.28 per 100,000 persons per year.

Perianal Keratinocyte Carcinoma

The MR for perianal KC was 0.23 per 100,000 persons per year for the period studied (males: 0.19, females: 0.25). Mortality rates for perianal KC increased by 21% (β = 0.011, CI: 0.007–0.016, p: .001). This trend was found in whites (β = 0.009, CI: 0.005–0.013, p: .003), but not in blacks (β = 0.009, CI: −0.003 to –0.020, p: .109). Perianal KC MRs were increasing for white women (β = 0.011, CI: 0.006–0.016, p: .002) and for black women (β = 0.009, CI: 0.003–0.014, p: .012), but not for men (β = 0.004, CI: −0.002 to 0.011, p: .149). For whites, MRs were significantly increasing in the southern states (β = 0.014, CI: 0.002–0.027, p: .026), but not in the northern states (β = .001, CI: −0.008 to 0.010, p: .796). Keratinocyte carcinoma MR for southern states was 0.23 and 0.22 for northern states for the period studied. Crude MR was 0.14 per 100,000 persons per year for patients younger than 65 years and for patients older than 85 crude MR was 1.42 per 100,000 persons per year.

Discussion

Deaths due to KC are increasing. Previous literature has reported that deaths due to KC have been decreasing through 2000. There was a 17% decrease in the KC MRs for years 1979–2000, followed by an increase of 44% for years 2000–2017 (Figure 2).3,4 The age-adjusted US MR was at 1.1 1980, dropped to 0.5 in 2000 and rose to 1.3 per in 2017 (per 100,000 per year). It is important to identify the causes for this trend, to evaluate our existing preventative and treatment methods. Possible causes are the increase of KC incidence, the increase of the immunosuppressed populations and changes in the accuracy of death certificates.

Figure 2.
Figure 2.:
Age-adjusted keratinocyte carcinoma mortality rates in the United States, 1979–2017. There is a 17% decrease in the KC MRs for years 1979–2000, followed by an increase of 44% for years 2000–2017. Rates are age-adjusted to the standard US 2000 population and expressed per 100,000 per year.

About half of the KC deaths were due to KC in chronically or intermittently sun-exposed areas with a striking increase in whites, but no change in blacks. The incidence of KC in light skin populations has been rising and possibly playing a role in the increase of the MRs.2 Exposure to UV radiation has increased, which is causing sun damage and hence increasing KC incidence.8 However, MRs were also increasing for whites, but not blacks for genital KC, suggesting that factors other than sun exposure may be playing a role.

Mortality rates may be affected by immigration. According to the Center for Immigration Studies, since 2000, darker skinned populations have increased rapidly. in the United States.8,9 The segment of the population with darker skin type has increased as a proportion of the total and this segment is at lower risk for KC.9

The human papillomavirus (HPV) has been associated with anal and genital KC10,11 We report increases in MRs of genital and perianal cancers. In the United States during 1999 to 2015, vulvar (1.3%) and anal SCC incidence increased for both men (2.1%) and women (2.9%).12 Increasing HPV vaccination will hopefully result in mortality reduction, but it is too soon now to see those effects. Penile SCC incidence rates have remained stable.12 The latest data on circumcision in the United States showed an overall increase in prevalence in males aged 14 to 59 years between 2000 and 2010.13 It has been demonstrated that circumcised men are at reduced risk of penile cancer.14 It could be suggested that the effect of circumcision was protective and hence men did not experience the increase in KC mortality like the women.

Increased use of immunosuppression treatments and worse prognosis of those patients may also be contributing to the observed MRs. Other factors possibly contributing to the observed mortality trends are increased use of immunosuppression treatments and worse prognosis of those patients.15,16 Long-term immunosuppression therapy, especially in solid organ transplant patients is a risk factor for SCC.16 Solid organ transplant patients develop more aggressive SCCs.17 Transplant recipients have higher skin cancer specific mortality than the general population, which is estimated at 35 per 10,000 person-years in the United States.18 Being black has been demonstrated to have lower chances for a transplant, higher propensity to be transplanted at a worse performing center and worse surgery outcomes.19,20 Furthermore, whites (compared to non-whites) have a hazard ratio of 10 for SCC-specific mortality in the post-transplant period.18 Those factors may be contributing to the increase in KC MRs in whites compared with blacks.

The noted trends in the MRs may also be affected by changes in the death certificate documentation. Other studies report that annual deaths from KC vary between 3,932 and 8,791 in the United States.7 Death certificates are indeed subject to error. They are completed by physicians who may not have cared for the deceased during the course of their illness. Those findings are calculated from various publications and not directly from death certificate records. Detailed examination of clinical records limited to patients of one hospital revealed large proportions of incorrect reports and even more inaccurate corresponding ICD-10 codes.21 This indicated potential underreporting SCC-specific mortality, although numbers of deaths studied were small and the sample may have been subject to selection bias by virtue of being restricted to one health system.21 Overreporting of SCC as a cause of death has also been documented: analysis of misclassified cases revealed that 85% of the cases coded as skin cancer of the scalp and neck were arising from mucosal sites in the head and neck, including larynx, pharynx, and oral cavity.22 These inaccuracies in death certificate documentation are the primary limitation of this study. More accuracy in documentation is needed to capture the magnitude of the nationwide KC burden in a population-based setting. We need more accurate death certificate documentation to capture the magnitude of the nationwide KC burden in a population-based setting.

Measuring and tracking KC mortality is of potentially great consequence for improving the health outcomes of our population. These dual imperatives define our current dilemma and we hope they can guide us to improved measurement of this consequential parameter. It is important to continue to study these parameters to determine which of factors is most relevant to our observations.

Dermatologic surgery, specifically Mohs Micrographic Surgery (MMS), is considered the gold standard treatment for KC with 99% cure rate.23 Therefore, the observed increase in MRs cannot be attributed in this treatment technique. However, immunosuppressed patients treated with MMS are at greater risk for postoperative complications, such as surgical site infection and wound dehiscence, which may increase mortality.24 Furthermore, findings consistently suggest that solid organ transplant patients do not demonstrate adequate sun-protection behaviors and do not think that their increased skin cancer risk is an important problem.25,26 A study evaluated sun-protection in Mohs healthy and immunosuppressed patients. Seventy-three percent of healthy and 74% of immunosuppressed participants identified sunscreen use as sun-protection. However, only 36% and 27%, respectively, used sunscreen daily.27 Therefore, it is critical that immunosuppressed patients receive systematic education about their increased SCC risk, the increasing KC mortality and effective sun protection behaviors. It should be highlighted that because they are at increased risk for aggressive SCC, visits to their dermatologist should be frequent to ensure prompt diagnosis.

References

1. Miller DL, Weinstock MA. Nonmelanoma skin cancer in the United States: incidence. Arch Dermatol 1994;30:774–8.
2. Rogers HW, Weinstock MA, Feldman SR, Coldiron BM. Incidence estimate of nonmelanoma skin cancer (keratinocyte carcinomas) in the U.S. Population, 2012. Arch Dermatol 2015;151:1081–6.
3. Lewis KG, Weinstock MA. Trends in nonmelanoma skin cancer mortality rates in the United States 1969 through 2000. J Invest Dermatol 2007;127:2323–7.
4. Wu W, Weinstock MA. Trends of keratinocyte carcinoma mortality rates in the United States as reported on death certificates, 1999 through 2010. Dermatol Surg 2014;40:1395–401.
5. CDC WONDER. Available from: https://wonder.cdc.gov/cmf-icd10.html. Accessed March 16, 2019.
6. Lewis KG, Weinstock MA. Nonmelanoma skin cancer mortality (1988–2000): the Rhode Island follow-back study. Arch Dermatol 2004;140:837–42.
7. Karia PS, Han J, Schmults CD. Cutaneous squamous cell carcinoma: estimated incidence of disease, nodal metastasis, and deaths from disease in the United States, 2012. J Am Acad Dermatol 2013;68:957–66.
8. Robinson JK, Rigel DS, Amonette RA. Trends in sun exposure knowledge, attitudes, and behaviors. J Am Acad Dermatol 1997;37:179–86.
9. Record 44.5 Million Immigrants in 2017. Available from: https://cis.org/Report/Record-445-Million-Immigrants-2017. Accessed March 25, 2019.
10. Dunne EF, Unger ER, Sternberg M, McQuillan G, et al. Prevalence of HPV infection among females in the United States. JAMA 2007;297:813–9.
11. Sichero L, Giuliano AR, Villa LL. Human papillomavirus and genital disease in men: what we have learned from the HIM study. Acta Cytol 2019;63:109–17.
12. Van Dyne EA, Henley SJ, Saraiya M, Thomas CC, et al. Trends in human papillomavirus-associated cancers - United States, 1999-2015. MMWR Morb Mortal Wkly Rep 2018;67:918–24.
13. Morris BJ, Bailis SA, Wiswell TE. Circumcision rates in the United States: rising or falling? What effect might the new affirmative pediatric policy statement have? Mayo Clin Proc 2014;89:677–86.
14. Morris BJ, Gray RH, Castellsague X, Bosch FX, et al. The strong protective effect of circumcision against cancer of the penis. Adv Urol 2011;2011:812368.
15. Organ Donor Statistics. Available from: https://www.organdonor.gov/statistics-stories/statistics.html. Accessed March 12, 2019.
16. Garrett GL, Lowenstein SE, Singer JP, He SY, et al. Trends of skin cancer mortality after transplantation in the United States: 1987 to 2013. J Am Acad Dermatol 2016;75:106–12.
17. Lanz J, Bouwes Bavinck JN, Westhuis M, Quint KD, et al. Aggressive squamous cell carcinoma in organ transplant recipients. JAMA Dermatol 2019;155:66–71.
18. Garrett GL, Blanc PD, Boscardin J, Lloyd AA, et al. Incidence of and risk factors for skin cancer in organ transplant recipients in the United States. JAMA Dermatol 2017;153:296–303.
19. Kilic A, Higgins RS, Whitson BA, Kilic A. Racial disparities in outcomes of adult heart transplantation. Circulation 2015;131:882–9.
20. Mustian MN, Shelton BA, MacLennan PA, Reed RD, et al. Ethnic and age disparities in outcomes among liver transplant waitlist candidates. Transplantation 2019;103:1425–32.
21. Wang DM, Besaw RJ, Karia PS, Schmults CD. Accuracy of death certification in cutaneous squamous cell carcinoma: a retrospective case review. J Am Acad Dermatol 2018;78:423–5.
22. Weinstock MA, Bogaars HA, Ashley M, Litle V, et al. Nonmelanoma skin cancer mortality. Arch Dermatol 1991;127:1194–7.
23. Skin Cancer Foundation. Mohs surgery: the most effective technique for treating common skin cancers. Available from: https://www.skincancer.org/treatment-resources/mohs-surgery/. Accessed September 10, 2019.
24. Basu P, Goldenberg A, Cowan N, Eilers R, et al. A 4-year retrospective assessment of postoperaative complications in immunosuppressed patients following Mohs micrographic surgery. J Am Acad Dermatol 2019;80:1594–601.
25. Walker K, Gardner K, Law A, Hawkins L, et al. Photoprotection knowledge and behaviours among organ transplant recipients. J Cutan Med Surg 2017;21:217–20.
26. Skiveren J, Mortensen EL, Haedersdal M. Sun protective behaviour in renal transplant recipients. A qualitative study based on individual interviews and the Health Belief Model. J Dermatolog Treat 2010;21:331–6.
27. Goldenberg A, Nguyen BT, Jiang SI. Knowledge, understanding and utilization of preventive strategies against nonmelanoma skin cancer in healthy and immunosuppressed Mohs surgery patients. Dermatol Surg 2014;40:93–100.
© 2020 by the American Society for Dermatologic Surgery, Inc. Published by Wolters Kluwer Health, Inc. All rights reserved.