TO THE EDITOR
Improvement in hematopoietic cell transplantation (HCT) strategies and supportive care have contributed to a growing number of long-term survivors. Cutaneous squamous cell carcinoma (cSCC) is the most common skin cancer in these HCT survivors, and there are reported associations with immunosuppressive therapies (e.g. calcineurin inhibitors), anti-fungals (e.g. voriconazole), and post-HCT complications such as chronic graft-versus-host disease (GVHD) (Kuklinski et al., 2017). That said, previous studies have included patients primarily treated during earlier transplant eras, and have not accounted for changing patterns in HCT practice such as: improved GVHD prophylaxis and management, greater utilization of non-myeloablative conditioning, and increasing numbers of HCTs performed in older individuals (D'Souza et al., 2020) who may be at higher risk of cSCC at baseline. Importantly, few studies have described the incidence and risk factors for de novo cSCC, and therefore have not differentiated new from recurrent disease.
We used a retrospective cohort design and followed the Strengthening the Reporting of Observational Studies in Epidemiology reporting guideline (Vandenbroucke et al., 2007) (Table S1) to describe the incidence of de novo cSCC (primary outcome) among contemporary cohort of adult (≥18 years) long-term (≥1-year) HCT survivors and evaluated the association between patient and HCT-related factors and cSCC. We included both autologous and allogeneic HCT survivors, allowing us to describe the burden of cSCC attributed to allogeneic HCT and/or associated exposures. Individuals with a history of cSCC prior to HCT were excluded. Our hypothesis was that allogeneic HCT survivors will have a greater risk of cSCC compared to autologous survivors, and that GVHD and/or its treatment will be a predictor of cSCC risk among allogeneic HCT survivors.
Cumulative incidence of de novo cSCC was calculated taking into consideration competing risk of death (Gray, 1988). Fine-Gray proportional sub-distribution hazard models were used to estimate the relationship between relevant covariates and cSCC risk. Hazard ratios (HR) and their 95% confidence intervals (CI) were determined to quantify magnitude of risk. Since the majority of survivors who developed cSCC were self-reported non-Hispanic white ([NHW] 92.6%) or older (≥50 years [79.0%]) at HCT, the analyses examining risk factors for cSCC were limited to survivors who were NHW and older at HCT. Variables included in the multivariable analyses were selected a priori, based on literature review; we evaluated collinearity between variables (threshold r≥0.5), and included the variable most likely to be associated with the biology of cSCC. We developed two multivariable regression models limited to: 1) older NHW survivors [N=754], 2) older NHW survivors of allogeneic HCT [N=317].
There were 1,867 patients who underwent HCT as adults at City of Hope between 2005 and 2014 and survived ≥1-year after HCT (index date); cohort characteristics are presented in Table 1. Median follow-up was 4.3 years (interquartile range [IQR: 2.6-6.6 years]), representing 9,084 person-years of follow-up. There were 81 HCT survivors who developed a de novo cSCC at a median of 3.3 years (IQR: 2.1-4.8 years) from HCT. The 5-year cumulative incidence of cSCC for the entire cohort was 5.1%; the incidence was highest for older NHW survivors (9.5%, Figure S1). The most common sites of involvement were the head (N=36, 44.4%) and extremities (N=35, 43.2%). Univariate analyses describing the association between patient and HCT-related variables and cSCC are presented in Table S2. In adjusted multivariable analyses, survivors of allogeneic HCT had a nearly four-fold (HR=3.60, 95% CI 2.41-7.71) risk of developing cSCC compared to autologous HCT survivors; Table 2. Among older NHW survivors of allogeneic HCT, history of moderate-severe skin GVHD within 1-year post-HCT was associated with a greater than two-fold (HR=2.54, 95% CI 1.05-6.12) risk of cSCC; Table 2.
Table 1.
Demographic and Treatment Characteristics of Hematopoietic Cell Transplantation (HCT) Survivors
| Total (N=1867) | Autologous (N=998) |
Allogeneic (N=869) |
|
|---|---|---|---|
| Median age at HCT (Q1, Q3), y | 54.1 (42.2, 61.4) | 56.5 (46.5, 63.0) | 51.1 (36.6, 59.3) |
| Age at HCT, No. (%) | |||
| <50 years | 737 (39.5%) | 321 (32.2%) | 416 (47.9%) |
| >=50 years | 1130 (60.5%) | 677 (67.8%) | 453 (52.1%) |
| Gender, No. (%) | |||
| Female | 794 (42.5%) | 407 (40.8%) | 387 (44.5%) |
| Male | 1073 (57.5%) | 591 (59.2%) | 482 (55.5%) |
| Race/ethnicity, No. (%) | |||
| White, Non-Hispanic | 1080 (57.8%) | 599 (60.0%) | 481 (55.4%) |
| White, Hispanic | 459 (24.6%) | 221 (22.1%) | 238 (27.4%) |
| Asian, any ethnicity | 210 (11.2%) | 97 (9.7%) | 113 (13%) |
| Black, African American | 90 (4.8%) | 72 (7.2%) | 18 (2.1%) |
| Other | 28 (1.5%) | 9 (0.9%) | 19 (2.2%) |
| HCT Conditioning, No. (%) | |||
| Non-myeloablative | 1042 (55.8%) | 576 (57.7%) | 466 (53.6%) |
| Myeloablative | 825 (44.2%) | 422 (42.3%) | 403 (46.4%) |
| Voriconazole exposure, No. (%) | |||
| No | 1506 (80.7%) | 951 (95.3%) | 555 (63.9%) |
| Yes | 361 (19.3%) | 47 (4.7%) | 314 (36.1%) |
| History of non-skin solid malignancy, No. (%) | |||
| None | 1746 (93.5%) | 940 (94.2%) | 806 (92.8%) |
| Any | 121 (6.5%) | 58 (5.8%) | 63 (7.2%) |
| Sirolimus >180 days, No. (%) | |||
| No | - | - | 368 (42.3%) |
| Yes | - | - | 501 (57.7%) |
| Tacrolimus >180 days, No. (%) | |||
| No | - | - | 289 (33.3%) |
| Yes | - | - | 580 (66.7%) |
| Acute GVHD, No. (%) | |||
| None/grade I | - | - | 474 (54.5%) |
| Grades II-IV | - | - | 395 (45.5%) |
| Chronic GVHD of the skin by 1-year, No. (%) | |||
| None | - | - | 359 (41.3%) |
| Any | - | - | 511 (58.8%) |
Abbreviations: Q1 = first quartile; Q3 = third quartile; y = years; No. = number; GVHD = Graft-vs-host disease
Table 2.
Multivariable Fine-Gray Proportional Sub-Distribution Hazard Models for Risk Factors for Squamous Cell Carcinoma (SCC) among Older (>=50 years) non-Hispanic White (NHW) autologous and allogeneic HCT survivors, and Older NHW allogeneic HCT survivors
| Older NHW (N=754) HR (95% CI) |
P-Value | Older NHW Allogeneic HCT (N=317) HR (95% CI) |
P-Value | |
|---|---|---|---|---|
| HCT Type | ||||
| Autologous | 1 [Reference] | - | - | |
| Allogeneic | 3.60 (1.83-7.07) | <0.001 | - | - |
| Gender | ||||
| Female | 1 [Reference] | 1 [Reference] | ||
| Male | 1.19 (0.71-1.99) | 0.517 | 1.01 (0.55-1.88) | 0.964 |
| HCT Conditioning | ||||
| Myeloablative | 1 [Reference] | 1 [Reference] | ||
| Non-myeloablative | 1.82 (0.90-3.69) | 0.096 | 1.41 (0.63-3.20) | 0.406 |
| History of non-skin solid malignancy | ||||
| None | 1 [Reference] | 1 [Reference] | ||
| Any | 1.53 (0.74-3.17) | 0.249 | 1.48 (0.65-3.35) | 0.349 |
| Voriconazole exposure, No. (%) | ||||
| No | 1 [Reference] | 1 [Reference] | ||
| Yes | 1.08 (0.58-2.01) | 0.816 | 0.88 (0.47-1.65) | 0.687 |
| Donor source | ||||
| HLA Matched Related | - | - | 1 [Reference] | |
| Other | - | - | 1.17 (0.64-2.13) | 0.615 |
| Chronic GVHD of the skin by 1 year | ||||
| None | - | - | 1 [Reference] | |
| Any | - | - | 2.54 (1.05-6.12) | 0.038 |
Abbreviations: HCT = Hematopoietic Cell Transplantation; HR = hazard ratio; CI = confidence interval; HLA = human leukocyte antigen; GVHD = Graft-vs-host disease
In this large and contemporary cohort of long-term survivors of HCT, the median time to diagnosis of de novo cSCC was 3-years from HCT, which coincides with a time when HCT survivors have largely transitioned to the community setting. Studies have shown that less than 20% of these long-term HCT survivors report routine skin cancer screening (Hahn et al., 2017). These low rates of screening are exacerbated by inconsistent national guidelines for skin cancer screening in HCT survivors (Mani et al., 2020). Our findings highlight the need for continued vigilance in these survivors, including those who are NHW and older at HCT. Improved access to screening and dermatology services (e.g. teledermatology) or technologies (e.g. digital capture and transfer of suspicious lesions to providers) (Jones et al., 2019) can optimize the care of survivors who are at highest-risk of developing skin cancers after HCT.
In our subset analyses, allogeneic survivors and those with a history of skin GVHD had an especially high-risk of developing cSCC. Previous studies have shown how local cytokines and chemokines interact with cellular toxicity mediators in the setting of skin GVHD, resulting in loss of epithelial and immune homeostasis (Weaver et al., 2013); immunosuppression may further exacerbate these alterations. Unlike studies in solid organ transplant recipients (Zamoiski et al., 2017), we found no differences in cSCC incidence by immunosuppressive drug class. This may be due to the shorter duration of immunosuppression in patients undergoing HCT, or evolving practice of HCT (e.g. decreased utilization of certain calcineurin inhibitors such as cyclosporine). Future studies may need to examine the attributable risk of prolonged immunosuppression after allogeneic HCT that is beyond the risk due to GVHD alone.
The findings from this study should be considered in the context of its limitations. Risk factors such as Fitzpatrick skin type or duration or type of ultraviolet light exposure were not reliably reported in the medical records, nor was extent of dermatology subspecialty follow-up; these were not accounted for in the analysis. Nevertheless, our study is one of the largest of its kind and provides much-needed guidance regarding risk-based screening for de novo skin cancers, taking into consideration the changing landscape of HCT practice. It also speaks to the importance of long-term surveillance for skin cancer, especially for the increasingly older survivors of allogeneic HCT or those with a history of skin GVHD, allowing for personalized prevention and screening strategies to decrease the burden of cancer-related morbidity over time.
Supplementary Material
ACKNOWLEDGEMENTS:
This study was supported, in part, by grants from the Leukemia and Lymphoma Society Scholar Award for Clinical Research (Armenian) and National Institutes of Health/National Cancer Institute (R01 CA249460 [Armenian, Geller]).
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
This study was presented, in part, at the 2019 annual meeting of the American Society of Hematology in Orlando, FL.
CONFLICT OF INTEREST:
FA: Research: Kwoya Kirin, Inc; Speakers Bureau: Mallinckrodt
SJF: Mustang Bio - Grant support, IP; Lixte Bio - Board Member
The remaining authors state no conflict of interest.
DATA AVAILABILITY:
Datasets related to this article can be found at http://dx.doi.org/10.17632/cfhtmjbycr.1, an open-source online data repository hosted at Mendeley Data (Berano Teh, Jennifer, 2020).
References
- D'Souza A, Fretham C, Lee SJ, Aurora M, Brunner J, Chhabra S, et al. Current Use and Trends in Hematopoietic Cell Transplantation in the United States. Biology of Blood and Marrow Transplantation 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gray RJ. A Class of K-Sample Tests for Comparing the Cumulative Incidence of a Competing Risk. The Annals of Statistics 1988; 16(3): 1141–54. [Google Scholar]
- Hahn T, Paplham P, Austin-Ketch T, Zhang Y, Grimmer J, Burns M, et al. Ascertainment of Unmet Needs and Participation in Health Maintenance and Screening of Adult Hematopoietic Cell Transplantation Survivors Followed in a Formal Survivorship Program. Biol Blood Marrow Transplant 2017;23(11):1968–73. [DOI] [PubMed] [Google Scholar]
- Jones OT, Jurascheck LC, van Melle MA, Hickman S, Burrows NP, Hall PN, et al. Dermoscopy for melanoma detection and triage in primary care: a systematic review. BMJ open 2019;9(8):e027529-e. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kuklinski LF, Li S, Karagas MR, Weng W-K, Kwong BY. Effect of voriconazole on risk of nonmelanoma skin cancer after hematopoietic cell transplantation. Journal of the American Academy of Dermatology 2017;77(4):706–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mani S, Khera N, Rybicki L, Marneni N, Carraway H, Moore H, et al. Primary Care Physician Perspectives on Caring for Adult Survivors of Hematologic Malignancies and Hematopoietic Cell Transplantation. Clinical Lymphoma Myeloma and Leukemia 2020;20(2):70–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Vandenbroucke JP, Von Elm E, Altman DG, Gøtzsche PC, Mulrow CD, Pocock SJ, et al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. PLoS Med 2007;4(10):e297. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Weaver CT, Elson CO, Fouser LA, Kolls JK. The Th17 pathway and inflammatory diseases of the intestines, lungs, and skin. Annu Rev Pathol 2013;8:477–512. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zamoiski RD, Yanik E, Gibson TM, Cahoon EK, Madeleine MM, Lynch CF, et al. Risk of Second Malignancies in Solid Organ Transplant Recipients Who Develop Keratinocyte Cancers. Cancer Res 2017;77(15):4196–203. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Datasets related to this article can be found at http://dx.doi.org/10.17632/cfhtmjbycr.1, an open-source online data repository hosted at Mendeley Data (Berano Teh, Jennifer, 2020).