If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The Presence of Betapapillomavirus Antibodies around Transplantation Predicts the Development of Keratinocyte Carcinoma in Organ Transplant Recipients: A Cohort Study
Department of Dermatology, Leiden University Medical Center, Leiden, The NetherlandsDepartment of Medical Microbiology, Leiden University Medical Center, Leiden, The Netherlands
Organ transplant recipients (OTRs) have an increased risk of developing keratinocyte carcinomas (KCs). The aim of this study was to correlate infection with human papillomaviruses (HPVs) belonging to the beta genus (Beta-papillomavirus (Beta-PV)) at transplantation with later development of KCs. In a cohort study, sera collected between 1 year before and 1 year after transplantation of OTRs transplanted between 1990 and 2006 were tested for antibody responses against the L1 capsid antigen of Beta-PV and other HPV genera (Gamma-, Mu-, Nu-, and Alpha-PV) using multiplex serology. The OTRs were followed for a maximum of 22 years. Cox regression models with KC, squamous cell carcinoma (SCC), and basal cell carcinoma (BCC) as outcome variables were used. Out of 445 OTRs, 60 had developed KC: 14 developed only SCC, 24 only BCC, and 22 both types of KC. The time-dependent hazard ratio (HR) to develop either or both types of KC, adjusted for age, sex, and transplanted organ, in tested Beta-PV-seropositive OTR around the time of transplantation compared with Beta-PV-seronegative OTR was 2.9 (95% confidence interval (CI) 1.3–6.4). The HR for SCC was 2.9 (95% CI 0.99–8.5) and for BCC it was 3.1 (95% CI 1.2–8.0). There was also an association between Mu-PV seropositivity and KC, but there were no significant associations between other HPV genera tested and KC. A positive seroresponse for Beta-PV around transplantation significantly predicted the development of KC in OTRs up to 22 years later, providing additional evidence that infection with Beta-PV has a role in KC carcinogenesis.
Abbreviations
BCC
basal cell carcinoma
CI
confidence interval
HPV
human papillomavirus
HR
hazard ratio
KC
keratinocyte carcinoma
LUMC
Leiden University Medical Center
OR
odds ratio
OTR
organ transplant recipient
PV
papillomavirus
SCC
squamous cell carcinoma
INTRODUCTION
Keratinocyte carcinomas (KCs) are the most prevalent malignancies seen in organ transplant recipients (OTRs). The incidence of squamous cell carcinoma (SCC) is 60–250 times increased compared with the general population, and for basal cell carcinomas (BCCs) this is 10–40 times (
Incidence and clinical predictors of a subsequent nonmelanoma skin cancer in solid organ transplant recipients with a first nonmelanoma skin cancer: a multicenter cohort study.
). In OTRs, SCCs are more frequently observed than BCC, with a SCC/BCC ratio of 5:1 compared with the general population where a SCC/BCC ratio of 1:4 is found (
Life-long immunosuppressive therapy is the most important risk factor for developing KCs in OTRs. Other important risk factors include sun exposure, male gender, older age, smoking, and fair skin with susceptibility to sunburn. These are similar risk factors as in the immunocompetent population (
). HPV types are classified in different classes (1–4) by the IARC (International Agency for Research on Cancer) based on their carcinogenic potential (
). Patients with epidermodysplasia verruciformis have an increased susceptibility to widespread Beta-papillomavirus (Beta-PV) infections of the skin that progress to SCC in one-third of the patients, mainly on sun-exposed sites (
). In the years after transplantation, OTRs start to develop a clinical picture that resembles that of patients with epidermodysplasia verruciformis. Epidemiological and experimental data have advocated a potential carcinogenic role of cutaneous HPV infection in skin cancer, especially of SCC in the immunosuppressed population (
), reflecting a past or present infection strong enough to evoke an immune response.
Most studies investigating the association between serological responses to HPV and SCC are cross-sectional or case–control studies that were carried out in the immunocompetent population, and showed an association between detection of Beta-PV antibodies and SCC (
). As far as we know, cohort studies investigating the association between serological responses to HPV and later development of a first KC were not carried out in OTR (
). Therefore, we designed a retrospective follow-up study to establish whether there is a relationship between the presence of serological responses to HPV, in particular Beta-PV, around transplantation and the development of KC (SCC and BCC) in the years after the organ transplantation.
RESULTS
The baseline characteristics of the 445 OTRs included in the study are shown in Table 1. The majority of patients were male (65%) and most patients (79%) underwent kidney transplantation. The mean age of the patients at transplantation was 47.3 years and the mean follow-up after transplantation was 11.8 year. A mean of 4.2 sera (range 1–17) was available from 1 year before to 1 year after transplantation. Among the patients who developed KC, the mean time from transplantation until the first KC was 7.7 years.
Table 1Baseline characteristics of the organ transplant recipients who were included in this study
All patients N(%)
Keratinocyte carcinoma
Squamous cell carcinoma, irrespective of basal cell carcinoma
Basal cell carcinoma, irrespective of squamous cell carcinoma
Supplementary Table S1 online shows the percentages of OTRs who were seropositive for any HPV type tested between 1 year before and 1 year after the transplantation, for all OTRs together and stratified for those with KC, SCC, and BCC. Table 2 summarizes Supplementary Table S1 online and provides the percentages of OTRs who were seropositive for at least one HPV type per HPV genus (Beta-, Gamma-, Mu-, Nu, or Alpha-PV). Seroresponses against the beta genus were more frequently observed in the patients with KC compared with patients without KC, reaching statistical significance (P=0.005; Table 2). Patients who were seropositive for at least one HPV type of a specific genus had a greater probability for being seropositive for another HPV genus (Supplementary Table S2 online).
Table 2Numbers and percentages of organ transplant recipients with positive antibody responses against one or more human papillomavirus (HPV) types of different HPV genera
The cumulative incidence of KC in Beta-PV-seronegative and Beta-PV-seropositive OTR is depicted in Figure 1 and Supplementary Figure S1 online. Log-rank tests were performed and showed a significant difference between Beta-PV-seropositive and -seronegative OTRs (P=0.043). The Kaplan-Meier plots for SCC and BCC showed similar curves as the plots for KC (Supplementary Figures S2 and S3 online). Table 3 shows the time-dependent nonadjusted hazard ratios (HRs) and the HRs adjusted for age at transplantation, sex, and type of transplantation for the development of KC, SCC, and BCC. A positive association was found between seropositivity for at least one Beta-PV type detected between 1 year before and 1 year after the organ transplantation and the development of KC, SCC, and/or BCC. After adjustment, Mu-PV seropositivity was also associated with KC (Table 3). The Cox proportional hazard analyses (Table 3) and the Kaplan-Meier plots of cumulative incidence of KC, SCC, and/or BCC in HPV-seronegative and -seropositive OTRs for the other HPV genera tested (Gamma-, Nu-, and Alpha-PV) showed no significant association between seropositivity against any of these HPV genera (Supplementary Figures S1–3 online).
Figure 1Kaplan–Meier plot for development of keratinocyte carcinomas (KCs) in relation to human papillomavirus (HPV) serology. The cumulative incidence of keratinocyte carcinomas (KCs) is statistically significantly higher in organ transplant recipients (OTRs) who were seropositive against one or more HPV types of the Beta-papillomavirus (Beta-PV) genus between 1 year before and 1 year after the transplantation (P=0.043).
Table 3Time-dependent Cox proportional hazard ratios for keratinocyte carcinoma, squamous cell carcinoma, and basal cell carcinoma in organ transplant recipients related to human papillomavirus (HPV) genus serostatus between 1 year before and 1 year after the transplantation
Keratinocyte carcinoma
Squamous cell carcinoma, irrespective of basal cell carcinoma
Basal cell carcinoma, irrespective of squamous cell carcinoma
Beta-PV species 1 (specifically HPV8 and HPV20) and Beta-PV species 3 (specifically HPV75 and HPV76) were most relevant for the positive association between Beta-PV seropositivity and the development of KC (Supplementary Figure S4 online). The potential importance of these Beta-PV types was confirmed for HPV8 and HPV75 by conditional step forward regression analyses, and a statistically significant negative association was observed for HPV96 (Table 4).
Beta-PV species 1 (HPV8 and HPV20) and Beta-PV species 3 (HPV76) were also associated with the development of SCC (Supplementary Figure S5 online), confirmed by conditional step forward regression analyses (Table 4). A statistically significant negative association was found for HPV24 and HPV38 (Table 4).
Table 4Association between 16 different Beta-papillomavirus (Beta-PV) types and skin cancer
Beta-PV species 2 (HPV15) and species 3 (HPV75 and HPV76) were relevant for the development of BCC (Supplementary Figure S6 online), which was confirmed in the conditional step forward regression analyses for HPV15 and HPV26, whereas there was a statistically significant negative association with HPV9 (Table 4).
DISCUSSION
We have found a statistically significant association between the development of KC, SCC, and BCC after transplantation and antibody responses to one or more Beta-PV types, measured in serum drawn in a window between 1 year before and 1 year after organ transplantation. The antibody response against Beta-PV types was specific for the beta genus, as there were no significant associations for positive seroresponses against the gamma, nu, and alpha genera with the development of KC, SCC, or BCC. Interestingly, we also found a positive association between KC and Mu-PV seropositivity, an HPV genus not recognized previously as potentially associated with KC development.
So far, most studies investigating HPV seroresponses in association with KC development concerned cross-sectional and case–control studies in the immunocompetent population (
). We are aware of only a few published cohort studies that were performed in the immunocompetent population. A Scandinavian study described a weak association for overall Beta-PV seropositivity before SCC diagnoses, whereas a significant association was observed for Beta-PV species 2 (odds ratio (OR)=1.3, 95% confidence interval (CI) 1.1–1.7), and for serum samples taken >18 years before diagnosis (OR=1.8, 95% CI 1.1–2.8) (
). A population-based cohort study in Australia revealed no associations between the presence of any of the Beta-PV antibodies and the occurrence of SCC (
A prospective pilot study of antibodies against human papillomaviruses and cutaneous squamous cell carcinoma nested in the Oxford component of the European Prospective Investigation into Cancer and Nutrition.
). One study investigated the possible role of seropositivity for different HPVs in the incidence of a subsequent KC in 107 patients with a follow-up of 5 years who were enrolled in a previous case–control study. HPV seropositivity at baseline was strongly associated with the risk of developing a second SCC after 5 years for a number of beta and gamma HPV types, but no association was found for developing a second BCC (
Seropositivity for human papillomavirus and incidence of subsequent squamous cell and basal cell carcinomas of the skin in patients with a previous nonmelanoma skin cancer.
). As far as we know, our study is the first cohort study with a prolonged follow-up period in OTRs in which the relationship between HPV serology before and just after the organ transplantation was studied in relation to later development of a first KC.
conducted a case–control study in OTRs from The Netherlands, United Kingdom, France, and Italy to investigate the association between SCC development and seroresponses against Beta-PV combined with detection of Beta-PV DNA in plucked eyebrow hairs. A positive association was found for concordant DNA and seropositivity for HPV36 (OR 2.4; CI 1.0–5.4), with similar, but not statistically significant, associations for HPV5, HPV9, and HPV24. Concordant Beta-PV DNA presence in hair and a serological antibody response for at least one HPV type was significantly associated with SCC risk (OR 1.6; CI 1.1–2.5). Another case–control study conducted in immunocompetent patients in Florida, USA, revealed a positive association between seropositivity to increasing numbers of Beta-PV types and SCC (OR 1.9; 95% CI 1.2–3.0) (
). A similar association for Beta-PV types DNA in eyebrow hairs was found in another clinic-based case–control study (4 types vs. HPV-negative: OR 2.0, 95% CI 1.1–3.8) (
performed a case–control study in an Italian immunocompetent population and found a significant 2- to 3-fold risk of SCC with seropositivity for any species 2 Beta-PV type (OR 3.3, 95% CI 1.2–8.7) and for the Gamma-PV types (OR 3.1, 95% CI 1.1–8.6). In a multinational case–control study in immunocompetent subjects, a positive antibody response against four or more Beta-PV types was associated with a doubled risk for SCC in Australia and the Netherlands (
Little data are published about the association between HPV infection and the development of BCC and the findings are contradictory. A relationship between seroresponses to Beta-PV and the development of BCC has been described previously (
reported a high frequency (11.5%) of HPV15 DNA in a small cohort of BCC patients, in agreement with the observed positive association between HPV15 and BCC in our study. However, some other studies did not report a positive association between Beta-PV seroreactivity and BCC (
). In a subgroup of patients in Florida, USA, with Beta-PV DNA-positive BCC, however, an association between a serological response against Beta-PV and BCC was observed (
). In a population-based case–control study of patients with SCC, BCC, and controls from New Hampshire, USA, it was found that seropositivity to Beta-PV (OR 1.5, 95% CI 1.0–2.1) was significantly associated with SCC. This risk increased with positivity for multiple HPV types. No associations were found with BCC risk (
confirmed these findings in a larger group of 2366 skin cancer cases and controls. We cannot exclude, however, that the OTRs who developed BCC in our study are at increased risk of developing SCC later on, because BCCs often precede the development of SCC in these patients.
Besides the advantages, the retrospective design of our study also has some disadvantages. For instance, the sera we tested were not systematically collected with the purpose to study a relationship with KC but were randomly selected from routinely archived sera collected for HLA antibody screening and other diagnostic purposes. The HRs we calculated were adjusted for sex, age at transplantation, and type of transplantation (
). We could not adjust for other risk factors for KC, like sun exposure, sunburns, skin type, and type of immunosuppression, because these data were not systematically collected for all patients. These risk factors are associated with an increased prevalence of HPV DNA in eyebrow hairs, but it is not known whether this is also the case for Beta-PV seroresponses (
A drawback of our study was that data about seroconversion were not systematically available. It can be speculated that seroconversion in the years following transplantation in OTRs who were seronegative around transplantation might have increased the risk of tumor development in the seronegative group that, if true, could have led to a stronger association between Beta-PV-type seropositivity and KC.
In general, a drawback of studies reporting the association between serological responses and skin cancer is that because of differences in the employed methodologies (initially starting with western blotting, followed by HPV-type-specific ELISAs using L1 virus-like particles, and more recently by high-throughput multiplex fluorescent bead–based assays enabling the simultaneous detection of antibody responses in large series of serologic samples against a variety of HPVs) the outcomes between the different studies cannot always be reliably compared. In our study, however, this is only true for the earlier studies referred to (western blot by
), as the other studies all used the same multiplex serology technique that was performed in the same laboratory in Heidelberg. Other factors that may explain the different associations between HPV infection and KC are differences in immunosuppressive regimens between centers that may exert different effects on the present HPV types, different exposure to sunlight reflected by the different latitudes of the centers, and differences in exposure to different HPV types.
The exact mechanism by which Beta-PV infection predisposes for the development of KC remains unclear. It has been hypothesized that the effects of the virus on UV-induced DNA damage withholds apoptosis and DNA repair by accumulation of mutations (
). This is supported by the fact that the E6 and E7 proteins from Beta-PV types are capable of inhibiting UV-induced cell cycle checkpoints and DNA repair mechanisms (
). The epidemiological evidence that HPV is involved in BCC carcinogenesis, however, is much weaker compared with the possible involvement of HPV in cutaneous SCC carcinogenesis. In conclusion, a detectable seroresponse for one or more Beta-PV types around the time of transplantation predicts an increased risk for the development of KC in OTR. This study provides additional evidence that infection with Beta-PV types has a role in keratinocyte carcinogenesis. Additional studies will be necessary to confirm the observed association between Mu-PV infection and KC development.
MATERIALS AND METHODS
Study design
Between 1966 and 2006, a total of 2,136 patients were transplanted in the Leiden University Medical Center (LUMC) (Supplementary Figure S2 online). OTRs consisted of both kidney transplant recipients and simultaneous pancreas–kidney transplant recipients.
During the past 25 years, sera were collected from a considerable number of OTRs. The most extensive source of sera consisted of samples collected from 1989 for clinical care–related screening and diagnostic purposes that were systematically stored in the Department of Medical Microbiology of the LUMC. In the scope of earlier case–control studies, additional sera had been collected and stored (
), pretransplant sera were retraced from the Eurotransplant serumbank. Altogether, 7,912 sera were available from 1,269 OTRs (Supplementary Figure S2 online). Because of practical reasons we had to limit the number of sera to be tested to a maximum of 3,000, the number thought to provide enough statistical power to answer our questions.
We selected 648 representative sera from all 164 OTRs who developed KC during follow-up for whom at least one serum sample was available. For each of these OTRs, we selected between 3 and 5 control OTRs without skin cancer. These patients were matched for type of transplantation (kidney transplant recipients or simultaneous pancreas–kidney transplant recipients), sex, and age at transplantation, resulting in the selection of 2,112 sera from 534 OTRs without skin cancer (Supplementary Figure S2 online). These sera had been collected between 7 years before and 42 years after the transplantation.
As we were specifically interested in the association between seroresponses to HPV around the transplantation and the later development of KC, we further restricted our selection to OTRs who had sera collected between 1 year before and 1 year after the transplantation. This resulted in a final cohort consisting of 1,880 sera from 445 OTRs, of whom 60 had developed KC (14 only SCC, 24 only BCC, and 22 both types of skin cancer) during the 22-year follow-up period. The number of sera collected in these patients ranged from 1 to 11 (mean: 4.2, median: 4.0). The remaining sera will be analyzed in a separate study.
Data about development of KC were retracted from the LUMC oncologic database, the LUMC pathology database, and by hand searching of the medical files. The study was approved by the medical ethical committee of the LUMC.
Serum processing and analysis
Serum samples were obtained from -20 °C freezers at the department of Medical Microbiology. They were defrosted and pipetted in 96-well plates. Some selected serum samples were excluded from further analysis, because either they could not be found, were not properly centrifuged initially, or contained too little volume (Supplementary Figure S7 online). The 96-well plates were shipped on dry ice to the German Cancer Research Center (DKFZ; Heidelberg, Germany) for analysis using Luminex technology (Austin, TX).
This multiplex serology technique is an antibody detection method based on glutathione S-transferase capture ELISA in combination with fluorescent bead technology. Positive serology cutoff points are standardized at 200 mean fluorescence intensity targeted on antibodies against major capsid antigen L1 of the various HPV types (
). Sera were tested for 7 Alpha-PV (HPVs 2, 3, 6, 7, 13, 16, and 27b), 16 Beta-PV (HPVs 5, 8, 9, 15, 17, 20, 23, 24, 36, 38, 49, 75, 76, 92, 93, and 96), 8 Gamma-PV (HPVs 4, 48, 50, 60, 65, 95, 101, and 103), 2 Mu-PV (HPV1 and HPV63), and one Nu-PV (HPV41).
The OTRs were considered seropositive for specific HPV genera or specific HPV types if at least one of the sera tested between 1 year before and 1 year after the transplantation had a positive antibody response against these specific HPV genera or specific HPV types.
Statistical analyses
Baseline characteristics are given in percentages and simple mean calculation. Kaplan–Meier survival analyses were used in conjunction with multivariable Cox proportional hazard methods to calculate corresponding HRs with 95% CIs for development of KC in relation to HPV serology. The date of transplantation was used as opening date. As censoring dates we used the date of diagnosis of first KC, the date of the patient’s death, the date of last follow-up, the date lost to follow-up, or the date of end of study (29 October 2012). Factors considered as potential confounders were sex, type of transplantation, and age at transplantation. The Cox models were adjusted for these variables. As the variable of HPV seropositivity had a time scope of 1year before and after the transplantation (starting point), time-dependent Cox regression analyses were performed. The analyses of SCC were performed regardless of the presence of BCC among the cases and controls and the analyses of BCC were performed regardless of the presence of SCC among the cases and controls.
To investigate whether there were Beta-PV types that were the driving force causing the positive association between Beta-PV seropositivity and the development of KC, we formed three subgroups: (1) patients who were seronegative for any Beta-PV type, (2) patients who were seropositive for the specific Beta-PV type regardless of the other Beta-PV types, and (3) patients who were seronegative for the specific Beta-PV type regardless of the other Beta-PV types. The cumulative incidence of KC, SCC, and BCC was calculated with Kaplan–Meier survival analyses. In addition, we performed conditional step forward logistic regression to get a rough estimation of which individual Beta-PV may possibly contribute to the association between seropositivity and the development of KC, SCC, and/or BCC. Starting with no variables in the model, all Beta-PV types, age, sex, and type of organ were introduced into the model, starting with the most relevant one until the addition of new variables did not improve the model any longer.
All analyses were performed with SPSS 20.0 software (IBM SPSS Statistics, IBM, Chicago, IL).
A prospective pilot study of antibodies against human papillomaviruses and cutaneous squamous cell carcinoma nested in the Oxford component of the European Prospective Investigation into Cancer and Nutrition.
Seropositivity for human papillomavirus and incidence of subsequent squamous cell and basal cell carcinomas of the skin in patients with a previous nonmelanoma skin cancer.
Incidence and clinical predictors of a subsequent nonmelanoma skin cancer in solid organ transplant recipients with a first nonmelanoma skin cancer: a multicenter cohort study.
37222-5&id=gr1.jpg){kind=link}