1. INTRODUCTION

Each year, malignancies, genetic diseases, hemoglobinopathies, and immunodeficiencies require hematopoietic cell transplantation (HCT) as potentially curative therapy for tens of thousands of patients. Eighty-five percent of patients who survive transplant-related complications in the first two years have a good prognosis for long-term survival (10 years) [1]. Female survivors are at a risk of developing chronic health conditions, including lower genital tract (LGT) precancer and cancer, premature menopause, and problems with sexual health. These women benefit from gynecologic care and follow-up post-HCT. An essential part of this care is screening for human papillomavirus (HPV)-related lower genital and cervical dysplasia/cancer, which includes cervical cytology and HPV testing and inspection of the LGT for HPV disease at pelvic exams. As in other populations, assessment of abnormalities noted on testing and biopsy of lesions enables early identification and treatment of premalignant lesions which, in turn, could decrease the risk of cervical and other LGT cancers in long-term survivors of allo-HCT [2].

HPV is the most common sexually transmitted infection, with a 27% estimated overall prevalence in women aged 14–59 years in North American and European populations [3,4]. The HPV prevalence increases from 20% in teenagers to 45% in women aged 20–24 years and declines to 5% in women over 50 years [5–7]. Ninety percent of HPV infections clear spontaneously within 2 years, but persistent HPV confers a significant risk of development of high-grade lesions and cancer. Additionally, HPV may reactivate and cause neoplasms later in life [8]. HPV-related disease typically takes 10 to 30 years to progress from initial epithelial changes to invasive cancer [9], illustrating how screening to identify and treat premalignant lesions in the general population can enable prevention of cervical cancer. Whether the time to these cancers is shorter in women post-HCT is not known.

Cervical cancer is the second most common cancer worldwide in women [10]. Vulvar and vaginal cancer are rare and account for only 7% of all gynecological malignancies [10] and HPV plays an important role in their development. High-risk HPV types (16, 18, 31, 33, 34, 45, 52, and 58) are responsible for 95% of cervical intraepithelial neoplasia (CIN) and invasive carcinomas [11,12] as illustrated by finding HPV genetic sequences in nearly all cervical cancers [12–14]. Primary vulvar and vaginal cancers are relatively rare in the LGT, but evidence of HPV DNA in precursor lesions (vulvar and vaginal intraepithelial neoplasia) has been reported with HPV 16 being the most common type detected [15]. Persistent infection with any of the oncogenic HPV types is essential for tumor cell transformation of cervical and other LGT cells [16–18]. Other immune compromised populations like HIV-infected women and those postrenal or other solid organ transplant are at risk of HPV-related disease throughout the LGTt.

HCT recipients may be at a higher risk of developing HPV-related neoplastic changes due to a number of factors. First, exposure to ionizing radiation and chemotherapy as part of treatment and conditioning regimens prior to HCT may lead to an increased risk due to their detrimental effects on immunity, making one more prone to infection or reactivation of HPV. Second, delayed immune recovery or prolonged immunosuppressive therapy (IST) for graft versus host disease (GVHD) increases the risk, as evidenced by an accelerated progression from cervical dysplasia to invasive carcinoma and a more aggressive disease course in immunocompromised patients [19]. Third, HPV and GVHD may be interrelated, as illustrated by a report of HPV reactivation in a post-HCT patient following use of topical immunosuppression and vaginal dilators [20]. Other gynecologic factors, like history of HPV-related disease and whether the woman has ever been and is currently sexually active also impacts the risk. Additionally, transplant-related factors related to the conditioning regimen and recovery including HPV vaccinations may alter this risk as well.

Given the growing literature on HPV and LGT cancers in recipients of HCT, many questions regarding the burden and risks of these complications have arisen in the practicing community. Herein, we conducted a systematic review of published peer-reviewed literature on the risk of HPV-related cervical and LGT precancer and cancer in HCT recipients. A meta-analysis could not be conducted due to the differences in both the selection of patients and the primary outcomes reported in various studies.

2. METHODS

A literature search was conducted on articles published from 1990 to 2018, indexed in the Cochrane Library, PubMed, Embase, Wiley Online Library.

3. RESULTS

Of the 344 articles identified, 27 provided LGT precancer and cancer information in posttransplant recipients. Removing redundancies and review articles, only 22 remained. Of these 22, five studies (two prospective cohort and three retrospective cohort studies) included cervical cytology results and examined the risk factors for cervical or LGT precancer. Thirteen retrospective cohort studies reporting the incidence of secondary cancers post-HCT included cervical cancer and six retrospective cohort studies reported on vulvar or vaginal cancer incidence.

4. RISK FACTORS AND MECHANISMS

Most studies to date in allo-HCT recipients have focused on cervical dysplasia rather than specifically on HPV-related LGT disease, and have shown that cervical HSIL and cervical dysplasia occur at a higher rate post-HCT than in general population. Women undergoing allo-HCT have an increased risk, due to chemotherapies and IST which come with the treatment of malignancy and cGVHD, respectively. The peak incidence for HPV disease appears to be years after HCT, suggesting HPV reactivation or new infection through a new sexual contact. Additionally, the conditioning regimen for HCT can potentially predispose the recipients to cervical dysplasia [33]. Finally, GVHD results in mucosal lesions involving genital tissues and may result in dysplasia either independently or synergistically with HPV involvement [49].

Besides quantitative risks, papers included in our review also identified that genital GVHD has various manifestations [50]. It can present as scarring or narrowing of the vaginal canal leading to shortening of the canal, with or without the presence of arcuate ridges or synechiae. It can also cause the formation of ulcers, or tender, inflamed erosions of the vulvar mucosa. Topical corticosteroid is the preferred treatment for mucocutaneous GVHD as systemic immunosuppression does not decelerate the course of genital GVHD progression. Topical estrogen may complement the effects of corticosteroids in some cases, as it promotes the growth of mucosa and may limit scarring [51]. But care must be taken, as IST and the usage of vaginal dilators has the potential to lead to widespread HPV infection [20].

HPV invokes both an antibody and cell-mediated response from the immune system, with evidence suggesting a central role of T-cells [52]. Posttransplant cell-mediated immunity is decreased, affecting T-cells, natural killer cells, and antibody production, which can be made worse by cGVHD and prolonged IST [53]. Cancer progression in active HPV infections can be affected by genetic factors, disease type, cGVHD, conditioning regimens, and treatments [54]. Aldabagh et al. [52] also highlighted an increased risk of multifocal HPV disease in immunocompromised women.

Since allo-HCTs carry a purportedly higher risk than auto-HCTs for developing new cancers post-HCT, these results present a challenge, as they collectively argue against this notion. The average risk described in allo-HCT recipients may be due to an insufficient follow-up period for cervical cancer, which has a long latency period, with some cases having been diagnosed after a decade. Secondary solid tumor development peaks at approximately 6.8 years post allo-HCT [47], and the incidence increases linearly over at least 20 years [37]. Longer follow up periods would allow the maximum influence of HCT-related factors to manifest on the cervix. All 13 studies on new cancers post-HCT lack HPV data (high risk versus low risk HPV infections) which can help identify which allo-HCT recipients are at an increased risk. Overall, the number of cervical cancer patients reported may have been underestimated, due to a significant fraction of children (<10 years at the time of HCT) included in the cohorts, who are unlikely to contract HPV during the latency period of cervical cancer. Rizzo et al. [37], Majhail et al. [39], and Kolb et al. [32] reported 14%, 6%, and 15% of total HCT recipients being under the age of 10, respectively. Ringdén et al. [30] and Bhatia et al. [33], did not specify the number, but did include young children in their cohorts. Additionally, the increased risk discovered by Bhatia et al. [33] and Shimada et al. [35] could be explained by their relatively small cohort sizes, where the small number of cervical cancers (four and six, respectively) could make the risk appear significantly elevated.

The exact relationship between GVHD and HPV reactivation and spread is uncertain. It is difficult to ascertain whether viral reactivation leads to GVHD or if GVHD and the IST associated with it, leads to HPV reactivation. IST hinders the ability of the immune system to clear HPV and increases the risk of dysplastic and neoplastic changes. Viral reactivation increases the likelihood of developing a pro-inflammatory microenvironment which promotes allo-immune activation, potentially leading to GVHD. The context of antigen presentation influences the immune response and the presence of self-antigens alone is enough to mount it [55]. Molecular patterns associated with pathogens and tissue damage are more likely to induce a pro-inflammatory response rather than anergy or an anti-inflammatory response [55]. Thus, when host self-antigens are presented to the donor immune system in this microenvironment, immune activation and local GVHD may become more likely. The need for the combination of both, allo-antigen and an inflammatory stimulus, to be sufficient for GVHD to occur is exhibited by the association of viral reactivation and GVHD in other mucosal environments. Sri et al. [20], described a 30-year-old woman, two years post-HCT for aplastic anemia, receiving systemic cyclosporine for cGVHD. Topical estrogen and corticosteroids, along with vaginal dilators were also used for the treatment of vaginal cGVHD. Condylomatous cervicitis was discovered upon colposcopy and biopsy for abnormal cervical cytology testing. Continued dilator therapy over the next 4 months resulted in the development of linear verrucous lesions in the vagina and vulva, which were then treated with laser therapy. Following the same patient, Buchan et al. [56] reported an outbreak of genital warts following the use of topical IST. This HPV reactivation limited the use of local IST and a vaginal estrogen ring was only able to delay vaginal scarring and necessitated a cruciate incision in the cervico-vaginal scar to relieve the hematocolpos that developed due to GVHD. This case highlights a mode of spread of HPV that might be easily missed. The use of local IST and dilator therapy for genital GVHD can enhance the spread of HPV, and could explain why vulvovaginal GVHD is a risk factor for cervical dysplasia in some papers. It also highlights the need for novel management of vaginal GVHD, and the interrelationship of GVHD and HPV.

5. ROLE OF HPV VACCINATIONS

Vaccination recommendations are provided in Table 4.

A limited number of HCT programs offer HPV vaccination post-HCT to provide protection to immunodeficient individuals against the pathogenic strains of HPV. Three vaccines are approved by the Food and Drug Administration to prevent HPV infection: Gardasil, Gardasil 9, and Cervarix. These vaccines are noninfectious, nonreplicating, subunits of viral-like particles. All three prevent infections with HPV types 16 and 18, two high-risk HPVs that cause about 70% of cervical cancers [57], 90% of anal cancers in men and women, 65% of vaginal cancers, and 78% of HPV-related vulvar cancers. Gardasil also prevents infection with HPV types 6 and 11, which cause 90% of genital warts. Gardasil 9 prevents infection with the same four HPV types plus five additional high-risk HPV types (31, 33, 45, 52, and 58).

For immunocompetent individuals, most guidelines recommend vaccinating females aged 9–14 years, before the onset of sexual activity, with a catch up period of 26 years. In many countries, a two-dose regimen is recommended for those under the age of 15 years with an interval of 6 months and a standard, three-dose regimen for those aged 15 years or older and the immunocompromised [58].

In the immunocompromised, including those who are HCT recipients, antibody titers are often lower when compared to the immunocompetent [59–62]. It was also noted that titers differ according to the immune therapy received, with mycophenolate producing lower HPV vaccine titers when compared to other drugs [63]. Nonetheless, these studies do not attribute any adverse effects to the HPV vaccine, nor do they alter the course of the original disease, demonstrating its safety [53,64,65]. Results from a recent large study on women posttransplant on low dose IST or not on any IST showed that they can mount a response to HPV vaccine and, thus, vaccination will likely become a recommended strategy in the future [66].

Female HCT recipients between the ages of 9 and 26 can ideally receive three doses of the HPV vaccine if there are no contraindications present. However, in sexually active women above the age of 26, vaccinations can be considered based on individual bases. Long-term safety and immunogenicity data is still lacking regarding the HPV vaccine post-HCT. Thus, further clinical research into understanding the course of LGT dysplasia, incidence of cancer, and the role of HPV is essential in guiding the establishment of HCT specific vaccination schedules in the future.

6. CHALLENGES

Due to the relative shortage of prospective trials on this topic and lack of relevant data, there are a number of challenges facing clinicians and researchers alike. A long-term follow-up period is needed to study the progressive changes associated with cervical precancer and cancer. Unfortunately, due to the age of the patients and the response to HCT, a significant number are lost to follow-up. Most of the reviewed publications describe their small cohorts as a limitation, warranting a large-scale study to be conducted. Although the Bethesda classification scale is used to identify the amount of dysplasia throughout these studies, there is a lack of consistency in the other variables being tested, which makes it extremely difficult to compile any meaningful numerical data, particularly the lack of pre and post-HCT HPV testing.

Cervical screening guidelines are provided in Table 5.

The timing and frequency of cervical screening in women undergoing HCT is still debatable, given the lack of prospective data. Frequent screening bears the risk of finding self-limiting, transient lesions which do not have any effect on treatment. The screening and management guidelines for cervical HPV disease in women post-HCT have not been formulated based on evidence. Instead they have mirrored what is advised for those with other immunodeficiencies. Established cervical screening programs for healthy populations require 3- or 5-year cervical cytology testing in sexually active females. However, this time frame is inappropriate for the immunocompromised patients and, additionally, the cytology testing alone may be insufficient for this group of patients. Based on individual risk factors, such as use of corticosteroids, duration of IST, time since coitarche and each woman's fertility goals, treatment of abnormalities should be individualized.

The same guidelines for cervical cancer screening and treatment in women with solid organ transplantation and HIV can be considered in post-HCT patients in the absence of substantial evidence. HPV-related disease screening starts earlier and occurs more frequently in HIV-positive women than in the general population, due to several reasons. Women infected with HIV have a greater risk of contracting high-risk HPV infections and CIN [67–69]. Sexually active adolescents who are HIV-positive have a higher rate of progression to abnormal cervical cytology [70]. These women have higher rates of vaginal, vulvar, and perianal neoplasia [71,72]. They also have higher rates of anal intraepithelial neoplasia and anal cancer, compared with the general population [73]. Abnormal cervical screening results and SIN lesions in HIV-positive women should be managed according to the ASCCP and The Panel on Opportunistic Infections guidelines and algorithms. Since these guidelines have recently changed, does it beckon a change in post-HCT cervical screening as well? In most of the articles reviewed, pre-HCT screening was generally ineffective in identifying women at an increased risk of developing cervical dysplasia, warranting the need for intense post-HCT monitoring. Cervical cancer screening is recommended every 1–3 years in HCT survivors from the ages of 21 to 65 years [74–76]; however, for nonsexually active females who have had three negative pap smears, discontinuation of screening may be considered on individual bases. Initiating annual gynecological screening is supported by the occurrence of HPV infection within the first few years after transplant, and in agreement with current transplant screening guidelines [53]. Women without extensive or genital GVHD, immunocompromised status, or malignancy relapse may be considered for longer screening intervals of 2–3 years if they have normal cytology testing in the first few years. Challenges also exist in finding the correct balance in treating GVHD and HPV. Genital HPV limits the use of local immunosuppression which may worsen scarring of the genital tract. As observed by Buchan et al. [56], only an estrogen ring was able to aid the healing of denuded vaginal mucosa as estrogen plays a role in the support and growth of mucosal surfaces [77], with limited effects on vaginal GVHD.

7. CONCLUSIONS

Our systematic review was conducted to compile relevant data regarding LGT precancer and cancer amongst HCT survivors and clearly suggests that good quality data regarding this topic are lacking and that further research is required. All studies show an increase in the amount of cervical dysplasia post-HCT, ranging from ASCUS to HSIL, with treatment-associated dysplasia occurring earlier in the course of the disease, and spontaneously resolving in some cases. The limited amount of progression data available shows either complete resolution of dysplasia or persistence until conization/hysterectomy. However, most of the published studies do not portray long-term outcomes of the complications or its interventions. Many allo-HCT recipients may require prolonged IST for GVHD and, thus, are at a higher risk for persistent HPV infections [22] and abnormal cervical smears [21]. Unrelated HLA-matched donor transplants are also associated with an increased risk for gGVHD among allo-HCT survivors [78] and, thus, confer an increased risk of cervical dysplasia [23]. Reports on new (secondary) cancer with LGT cancer data, following cohorts of allo and auto-HCT recipients, do not report consistent results. There is a lack of large prospective studies that follow post-HCT women to study the progression of LGT dysplasia to cancer.

Additionally, due to the absence of postvaccination efficacy studies, and to increased risk of disease from non-HPV vaccine strains, patients require intense gynecologic care post-HCT with dedicated gynecologists as part of survivorship program, and most countries recommend more frequent screening for these women compared to the general population [79–81]. Further research on this topic will help clarify screening measures and risk factors to effectively identify high-risk individuals and preemptively uncover any high-grade cervical precancer and cancer.

CONFLICTS OF INTEREST

None relevant.

AUTHORS' CONTRIBUTIONS

MST and SKH designed the study. All authors played a significant role in each step of manuscript writing and vouch for the accuracy and contents of the manuscript. All authors approved the final version of the draft.

DISCLOSURES

SKH: Honorarium Mallinckrodt, Janssen, Novartis, Pfizer