Beyond Prophylaxis: Could the HPV Vaccine be Repurposed in Skin Cancer Treatment

Rachel Sally [1]
[1] New York University Grossman School of Medicine, New York
Correspondance: Rachel.Sally@nyulangone.org

---

Counseling patients—in pediatrics, sometimes their parents—on human papillomavirus (HPV) vaccination frequently includes a turn of phrase believed to be so persuasive that not only is there an entire CDC webpage dedicated to it, but it has practically become a commandment on infographics for patients on cancer prevention—one has only to peruse WebMD for moments to find dozens of examples (1, 2). Medical students across the country are taught to lean in and say, “this is one of two vaccines that can prevent cancer.”

Papillomaviridae is a family of mucosal and cutaneous epitheliotropic viruses that cause hyperproliferative lesions. For over four decades, the causative link between high risk (HR) mucosal HPV and malignancies, particularly cervical cancer, has been robustly established. HPV is currently the most common sexually transmitted infection worldwide and represents a tremendous disease burden, with up to 5% of cancer diagnoses attributable to HPV infection (3, 4). Hence the importance of counseling on the vaccines (bivalent, quadrivalent, and nonavalent) that exist against a number of HR HPV types in order to prevent both benign and malignant diseases. The high risk mucosal types of HPV all fall within the alpha genus of papillomaviridae. Their mechanism of mucosal carcinogenesis is dependent on continuous expression of oncoproteins E6 and E7, often following a DNA integration event (5). During HR HPV chronic infection, these oncoproteins interact with various cellular targets to prevent apoptosis and promote unchecked growth.

Viruses from the beta, gamma, mu, and nu genera (and some alpha HPV types), colonize the epidermis, with hair follicle stem cells representing a natural reservoir of persistent infection (6). Cutaneous HPVs are ubiquitous on human skin and have been increasingly implicated in the development of non-melanoma skin cancer (NMSC), particularly 𝛽-HPVs and squamous cell carcinomas (SCC). However, up to 90% of healthy individuals test positive for cutaneous 𝛽-HPVs (5). In contrast to mucosal HR HPV carcinogenesis, it appears that E6 and E7 expression is only required at the beginning of skin carcinogenesis. The expression of these oncoproteins facilitates the accumulation of UV-induced mutations by preventing apoptosis in skin cells with high mutation burdens, known as the “hit-and-run” mechanism (5). The high prevalence of HPV in healthy people combined with a lack of evidence for an integration event as in mucosal carcinogenesis and an absence of active virus expression in cutaneous tumor samples support the hypothesis of synergism between UV radiation and HPV in the early pathogenesis but not maintenance of cutaneous SCC (7).

Currently, the FDA-approved HPV vaccines contain virus-like particles (VLP) of L1 major capsid proteins. This protein is not conserved among HPV types, with only 60% sequence homology between 𝛼- and 𝛽-types, and the resulting antibodies after vaccination are thus theoretically highly type-specific to 𝛼-HPV (8, 9). Though these vaccines were developed to prevent mucosal infection, there have been numerous reports of complete resolution of disseminated, treatment-resistant verruca vulgaris after quadrivalent HPV vaccination, even when the HPV type isolated from the lesions did not match the quadrivalent vaccine, which covers types 6, 11, 16, and 18, suggesting the existence of cross-protection against heterologous types (10).

Further, there exist in the literature five recent reports of a novel use of the quadrivalent and nonavalent HPV vaccines in patients with cutaneous squamous cell carcinoma. Nichols et al. offered two patients with SCC and BCC three systemic doses of the quadrivalent vaccine for prophylaxis for skin cancer. Sixteen months after the first dose, the male patient’s new cancerous lesion rate had a 62.5% reduction in new SCC (12 to 4.4 per year) and 100% reduction in BCC (2.25 to 0) and at 13 months, the female patient had a 66.5% reduction in SCC (5.5 to 1.84) and 100% reduction in BCC (0.92 to 0) (11).

The third case is the first documented therapeutic use of HPV vaccination and involves a 90-year-old immunocompetent patient with multiple inoperable cutaneous squamous cell carcinomas, biopsy-proven to be basaloid SCC, a rare SCC characterized by invasive growth, high recurrence rate after resection, and a propensity for metastasis. The patient underwent Mohs surgery on the largest tumor but given the severity of the tumor burden and the patient’s advanced age, additional surgery and radiation were deemed infeasible; the patient subsequently declined systemic chemotherapy. She was treated with a series of systemic and intratumoral nonavalent HPV vaccines (2 doses intramuscularly, 4 doses intratumorally in the largest 3 tumors over 10 months) to complete resolution of her tumors, including the non-injected ones. At eleven months after the first intratumoral dose, a small papule at the site of a previous large tumor was biopsied and showed no histologic evidence of residual disease. At 24 months after her first intratumoral dose, the patient remained with a complete response to treatment, with no clinical evidence of SCC (12). The same research team performed the treatment—two intramuscular and two intratumoral injections with the nonavalent vaccine—on an 87-year-old immunosuppressed renal transplant recipient with inoperable SCC in situ who elected to forego radiation. His lesion resolved with biopsy-proven histologic cure (13). The final case currently reported is from Geizhals and Lebwohl, who treated an 84-year-old immunocompetent patient with multiple invasive cutaneous keratoacanthoma-type SCC with 2 intramuscular and 3 intratumoral injections of the nonavalent vaccine. Ten months after the first injection, there was neither clinical nor histologic evidence of residual carcinoma (14).

The current understanding of the cofactor-like hit-and-run relationship between cutaneous 𝛽-HPV and early squamous cell carcinogenesis would not suggest cross-protectivity and immunogenicity of highly 𝛼-type-specific L1-based vaccines for established SCC. However, the growing body of observational data of the efficacy of the vaccine as a therapeutic treatment for a number of cutaneous HPV-related conditions, from verruca vulgaris to, rather strikingly, aggressive squamous cell carcinomas implies that the reality of immunologic treatment based on specific HPV serotypes is complicated.

As with any case report, there are significant limitations to their conclusions, not least that there are currently only 5 patients with cutaneous carcinomas known to be treated with this modality. A major question stemming from the lack of controls is whether the possible immunogenic effects of the vaccines were due to the VLP contents or the vaccine adjuvants. While regression of non-injected tumors suggests systemic effects beyond local adjuvant augmentation, a broader downstream adjuvant-activated immunologic cascade cannot be ruled out. The lack of HPV testing is also a hindrance to broader conclusions being drawn from these reports, as understanding cross-protectivity is rendered impossible without knowing the serotypes present. Another distinct possibility, particularly given the small sample size, is that the observed regressions were spontaneous.

There are a number of ongoing investigations aimed at creating papillomavirus vaccines based on the more broadly conserved but less immunogenic L2 minor capsid proteins to broaden protection to include heterologous HPV types, including cutaneous (15). However, these anecdotal studies raise the possibility of an already available alternative to the current orthodoxy of physical destruction (resection, ablation, topical/intralesional chemotherapeutics) for treatment of HPV-related cutaneous SCC. Further exploration and controlled experimentation is warranted into whether the commercially available HPV vaccines may be used as therapeutic, not just prophylactic, cancer vaccines. More robust and durable data must be collected and validated, and several questions must be answered, the most important being what the immunologic mechanism of action is; whether there are predictors of outcomes such as HPV vaccination status, serotype, and tumor type; and the safety profile of using the vaccine as a therapeutic agent.

DISCLOSURES
Funding: None.
Conflicts of interest: None.
Availability of data and materials: Not applicable.
Authors’ contributions: Authors listed in the manuscript have contributed per submission guidelines and standards for authorship.
Code availability: Not applicable.
Ethics approval: Not applicable.
Consent to participate: Not applicable.

REFERENCES
1.     Vaccines (Shots) Against Cancer | CDC. Published December 17, 2020. Accessed July 26, 2021. https://www.cdc.gov/cancer/dcpc/prevention/vaccination.htm

2.     Can You Prevent Cancer? What Helps Lower Your Risk. Accessed July 26, 2021. https://www.webmd.com/cancer/cancer-prevention-strategy

3.     Brianti, P., De Flammineis, E., & Mercuri, S. R. (2017). Review of HPV-related diseases and cancers. The new microbiologica, 40(2), 80–85.

4.     Rosalik, K., Tarney, C., & Han, J. (2021). Human Papilloma Virus Vaccination. Viruses, 13(6), 1091. 

5.     Gheit T. (2019). Mucosal and Cutaneous Human Papillomavirus Infections and Cancer Biology. Frontiers in oncology, 9, 355.

6.     Köhler, A., Forschner, T., Meyer, T., Ulrich, C., Gottschling, M., Stockfleth, E., & Nindl, I. (2007). Multifocal distribution of cutaneous human papillomavirus types in hairs from different skin areas. The British journal of dermatology, 156(5), 1078–1080.

7.     Arron, S. T., Ruby, J. G., Dybbro, E., Ganem, D., & Derisi, J. L. (2011). Transcriptome sequencing demonstrates that human papillomavirus is not active in cutaneous squamous cell carcinoma. The Journal of investigative dermatology, 131(8), 1745–1753.

8.     Yadav, R., Zhai, L., & Tumban, E. (2019). Virus-like Particle-Based L2 Vaccines against HPVs: Where Are We Today?. Viruses, 12(1), 18.

9.     de Villiers, E. M., Fauquet, C., Broker, T. R., Bernard, H. U., & zur Hausen, H. (2004). Classification of papillomaviruses. Virology, 324(1), 17–27.

10.   Cyrus, N., Blechman, A. B., Leboeuf, M., Belyaeva, E. A., de Koning, M. N., Quint, K. D., & Stern, J. J. (2015). Effect of Quadrivalent Human Papillomavirus Vaccination on Oral Squamous Cell Papillomas. JAMA dermatology, 151(12), 1359–1363.

11.   Nichols, A. J., Allen, A. H., Shareef, S., Badiavas, E. V., Kirsner, R. S., & Ioannides, T. (2017). Association of Human Papillomavirus Vaccine With the Development of Keratinocyte Carcinomas. JAMA dermatology, 153(6), 571–574.

12.   Nichols, A. J., Gonzalez, A., Clark, E. S., Khan, W. N., Rosen, A. C., Guzman, W., Rabinovitz, H., Badiavas, E. V., Kirsner, R. S., & Ioannides, T. (2018). Combined Systemic and Intratumoral Administration of Human Papillomavirus Vaccine to Treat Multiple Cutaneous Basaloid Squamous Cell Carcinomas. JAMA dermatology, 154(8), 927–930.

13.   Nichols, A. J., De Bedout, V., Fayne, R. A., Burke, G. W., Kirsner, R. S., & Ioannides, T. (2020). Systemic and intratumoral 9-valent human papillomavirus vaccine treatment for squamous cell carcinoma in situ in a renal transplant recipient. JAAD case reports, 6(4), 289–291.

14.   Geizhals S, Lebwohl MG. (2020). The Successful Treatment of Multiple Cutaneous Malignancies with HPV Vaccination: Case Report. J of Skin, 4(2), 148.

15.   Vinzón, S. E., Braspenning-Wesch, I., Müller, M., Geissler, E. K., Nindl, I., Gröne, H. J., Schäfer, K., & Rösl, F. (2014). Protective vaccination against papillomavirus-induced skin tumors under immunocompetent and immunosuppressive conditions: a preclinical study using a natural outbred animal model. PLoS pathogens, 10(2), e1003924.

How to Define Death: Variation in Donation After Circulatory Death Policies

Devon E. Cassidy [1], Meredith Barrett [2], Michael J. Englesbe [2], Valeria S. M. Valbuena [2, 3, 4]
[1] University of Michigan Medical School, Ann Arbor, Michigan
[2] University of Michigan Medical School Department of Transplantation Surgery, Ann Arbor, Michigan
[3] University of Michigan Medical School Centers for Outcomes and Policy, Ann Arbor, Michigan
[4] National Clinician Scholars Program
Correspondance: CDevon@med.umich.edu

---

ABSTRACT
Organ transplantation utilizes a shared and scarce resource. In order to best utilize this resource, a network of organ procurement organizations, hospitals and individuals must work together. National societies make recommendations for policies that govern organ transplantation recoveries, however at each tier of the network there is room for variability. Donation after circulatory death (DCD) policies are one example of organ transplantation policies that are not standardized.  The American Society of Transplantation Surgeons defines death in DCD recoveries as “irreversible cessation of cardiac and respiratory function.” However, many individual hospitals have policies that may differ from this practice of observing pulseless electrical activity (PEA) for the ASTS recommended wait time of 2 minutes. In this study, we examined the DCD protocols of 50 adult hospitals representing a single OPO within Michigan. We hypothesized there would be institutional variance in the definition of death, the provider who can declare death and maximum wait time for the donor to expire after extubation until organ recovery is no longer pursued. We found that there was substantial variation in how each hospital defined death, with the most common definition being asystole. Most hospitals require a physician to declare death in DCD and the minutes to expire range from 60 to 120 minutes. Given that the difference between PEA and asystole may result in time lost and organs to become nonviable, we recommend that standard policies are created and there is increased education to physicians and designees that declare death in DCD recoveries. 

---

INTRODUCTION
Although hospitals have policies outlining their protocol for Donation after Circulatory Death (DCD) organ donation, the rules and regulations around withdrawal practices vary significantly (1). National societies including the American Society of Transplant Surgeons (ASTS) make recommendations for DCD recoveries, and organ procurement organizations provide guidelines to hospitals within their region, but these recommendations are open to institutional interpretation (2, 3). Given the importance of timely recovery of organs for transplantation, the variation in hospital DCD policies should be analyzed to limit organ discard. 

There are many aspects of DCD protocols which can be scrutinized, but one specific area of interest is how hospital policies define death to allow for organ recovery. Death in DCD recoveries is defined by the ASTS as “irreversible cessation of cardiac and respiratory function” which in practice has come to mean observing pulseless electrical activity for a predetermined wait time to insure autoresuscitation does not occur (2). However, the power to write DCD policies lies within individual hospitals, and little is known about how closely these policies follow society guidelines.

METHODS
Within this context, we sought to explore the variation in how hospital policies define death to allow for organ recovery. Additionally, we sought to investigate variation in the providers who declare death and the maximum time between extubation and death. Using content analysis, we examined 50 Michigan adult hospitals DCD policies, all serviced by a single organ procurement organization. We hypothesized there would be substantial variation in the definition of death within Michigan hospitals.

Hospital DCD policies were identified through Gift of Life Michigan. The study was submitted to the university Institutional Review Board and met the criteria for exemption from further IRB oversight. Written policy documents were accessed using iTransplant database in August, 2021 utilizing the most up to date policy that was provided. A sample representing hospitals with diversity in location, size, and type (transplant center non-transplant center) was used. Content analysis was conducted for the hospital policies (4). The protocols were manually searched and coded for the presence or absence of a definition of death. Policies were coded by a single investigator (D.C.) and analyzed for observation of ASTS guidelines.

RESULTS
There was substantial variation in how each policy defined death, with the most common definition being asystole. Figure 1 demonstrates the different definitions of death encountered in the hospital DCD policies and the number of hospital policies that had these definitions. Of note, policies often contained more than one definition of death in their DCD protocol. Although ASTS recommends waiting for the onset of pulseless electrical activity, this was only found in 46% of policies. Eight policies did not clearly define the declaration of death. Figure 2 illustrates the process of DCD recoveries and how the maximum time from extubation to expiration ranged from 60 to 120 minutes in this study. Of note, 14% of hospitals allowed the recovery team to set the maximum time to expire and 14% did not provide a time within their policy. Additionally, we found that 96% of hospitals required a physician to declare death in DCD, in contrast to a nurse or advanced practice provider.

Figure 1: Michigan hospital policy definitions of death.

Figure 2: Donation after circulatory death timeline from withdrawal procurements.

DISCUSSION
There is broad variation in the definition of death across hospitals in Michigan. This is just one aspect of many components of DCD donation that can result in delays in declaration and recovery. Given this variation in policy—or lack of documented policy at all—there are likely donors which are not declared dead until asystole (flatline of the ECG) is noted, which may jeopardize the donation intentions of the family.

This study has several limitations, including that only written DCD hospital policies were reviewed. These policies may not accurately reflect actual performance, may be outdated, and practices at the selected hospitals may not reflect hospital policy. Additionally, this study utilized a single coder and changes may have been made to polices since the data abstraction period. Finally, a diverse selection of the hospitals in Michigan was used, so conclusions about statewide policy variation should be made with caution.

In order to optimize organ utilization, a stakeholder-driven standardization of DCD policies is necessary. Beginning at an institutional level, it will be key to examine and maintain hospital DCD policies that align with current national consensus practice guidelines. Including stakeholders such as donor families, organ procurement organizations, hospital ethics representatives, hospital leadership, transplant centers, nurses, and anesthesia providers must be engaged in this work. The ASTS and the Association of Organ Procurement Organizations have put forward best practices for DCD liver recovery that serve as a model (5). There is a shared goal amongst stakeholders to maximize the gift of organ donation and a broad adaptation to a shared policy will allow for improved outcomes for donors and recipients alike.

DISCLOSURES
Funding: Not applicable.
Conflicts of interest: None.
Availability of data and materials: Not applicable.
Code availability: Not applicable.
Ethics approval: Not applicable.
Consent to participate: Not applicable.

REFERENCES
[1] Rhee, J.Y., Ruthazer, R., O’Connor, K., et al. (2011). The impact of variation in donation after cardiac death policies among donor hospitals: a regional analysis. American Journal of Transplantation, 11(8), 1719-1726. doi:10.1111/j.1600-6143.2011.03634.x

[2] Reich, D.J., Mulligan, D.C., Abt, P.L., et al. (2009). ASTS recommended practice guidelines for controlled donation after cardiac death organ procurement and transplantation. American Journal of Transplantation, 9(9), 2004-2011. doi:10.1111/j.1600-6143.2009.02739.x

[3] Choubey, A.P., Siskind, E.J., Ortiz, A.C., et al. (2020). Disparities in DCD organ procurement policy from a national OPO survey: A call for standardization. Clinical Transplantation, 34(4). doi:10.1111/ctr.13826

[4] Bengtsson, M. (2016). How to plan and perform a qualitative study using content analysis. NursingPlus Open, 2, 8-14. doi:10.1016/j.npls.2016.01.001

[5] Hobeika, M.J., Glazner, R., Foley, D.P., et al. (2020). A Step toward Standardization: Results of two National Surveys of Best Practices in Donation after Circulatory Death Liver Recovery and Recommendations from The American Society of Transplant Surgeons and Association of Organ Procurement Organizations. Clinical Transplantation, 34(10), e14035. doi:10.1111/ctr.14035