Abstract
Purpose
Although cervical squamous cell carcinoma (SCC) and adenocarcinoma (AC) are both caused by human papillomavirus (HPV) infection, they differ in cofactors such as cigarette smoking. We assessed whether these cofactor differences translate into differences in second cancer risk.
Patients and Methods
We assessed second cancer risk among 85,109 cervical SCC and 10,280 AC survivors reported to population-based cancer registries in Denmark, Finland, Norway, Sweden, and the United States. Risks compared to the general population were assessed using standardized incidence ratios (SIR).
Results
Overall cancer risk was significantly increased among both cervical SCC survivors (n = 10,559 second cancers; SIR, 1.31; 95% CI, 1.29 to 1.34) and AC survivors (n = 920 second cancers; SIR, 1.29; 95% CI, 1.22 to 1.38). Risks of HPV-related and radiation-related cancers were increased to a similar extent among cervical SCC and AC survivors. Although significantly increased in both groups when compared with the general population, risk of smoking-related cancers was significantly higher among cervical SCC than AC survivors (P = .015; SIR for cervical SCC = 2.07 v AC = 1.78). This difference was limited to lung cancer (SIR for cervical SCC = 2.69 v AC = 2.18; P = .026). The increased lung cancer risk among cervical AC survivors was observed for both lung SCC and lung AC. SIRs for second cancers of the colon, soft tissue, melanoma, and non-Hodgkin's lymphoma were significantly higher among cervical AC than SCC survivors.
Conclusion
The second cancer profiles among cervical SCC and AC survivors mirror the similarities and differences in cofactors for these two histologies. Because smoking is not a cofactor for cervical AC, the increased lung cancer risk suggests a role for additional factors.
INTRODUCTION
Cervical cancer is the second most common cancer among women, with an estimated worldwide burden of 493,000 new cases and 274,000 deaths each year.1 Cervical cancer presents in two major histologic forms, namely, squamous cell carcinoma (SCC), which constitutes 80% to 85% of all cases, and adenocarcinoma (AC), which constitutes 15% to 20%.2 Cervical SCC and AC are both caused by human papillomavirus (HPV) infection and share several risk factors, including high-risk sexual behavior.3–7 However, the cofactor profiles of SCC and AC also show important differences.3–7 The most prominent difference pertains to cigarette smoking, which is associated with increased risk of SCC, but not AC.3–7 Although high parity is associated with increased risk of both histologic subtypes, the association is believed to be weaker for AC than SCC.4,5,8 Obesity is associated with increased risk of AC, but not SCC,9 and patients with cervical AC have been shown to be of relatively higher socioeconomic status (SES) than those with SCC, in some,10 but not all studies.11,12
Previous studies have quantified the increased risk of second primary cancers among cervical cancer survivors.13–17 This increased risk arises from several factors, including radiation treatment, HPV infection, cigarette smoking, and hormonal factors.13–17 However, no study has assessed second cancer risk separately for cervical SCC and AC survivors. Because cofactors such as cigarette smoking, which contributes to second cancer risk, differ between SCC and AC, it is important to assess whether these cofactor differences translate into differences in second cancer risk. In a previous analysis, we assessed long-term risk of second primary cancers among 104,760 cervical cancer survivors who were observed for more than 40 years.13 In this study, we assessed second cancer risk separately for cervical SCC and AC survivors.
PATIENTS AND METHODS
The study cohort comprised women diagnosed with invasive cervical cancer reported to population-based cancer registries in Denmark (1943 to 1998), Finland (1953 to 2001), Norway (1953 to 1999), Sweden (1958 to 2001), and nine areas in the United States covered by the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) program (1973 to 2002).13 Follow-up for invasive second primary cancers began 1 year after cervical cancer diagnosis and ended at the earliest occurrence of second cancer diagnosis, death, or end of study period. Incident second primary cancers were identified by linkage within the respective cancer registry.13
Using histology codes noted in Appendix Table A1, we classified 104,760 cervical cancers into SCC (n = 85,109) and AC (n = 10,280). Women with other/unknown histologies (n = 9,371) were excluded. We were unable to separate adenosquamous histologies from ACs for the non-SEER registries; however, for SEER program registries, adenosquamous cases were included in the other/unknown histologies category and were thus excluded.
Separately for SCC and AC, we compared risk for each second cancer with that in the general population using the standardized incidence ratio (SIR), calculated as the ratio of observed and expected number of cancers. The expected number of cancers was calculated by applying female general population cancer incidence rates, specific to registry, age (in 5-year groups), calendar year (in 5-year groups), and race (only for SEER data), to person-years accrued by cervical cancer survivors. Exact Poisson 95% CIs were calculated for the SIR estimates.18
We grouped second cancer sites as smoking-related cancers (lip, tongue, mouth, pharynx, esophagus, larynx, trachea/bronchus/lung, pancreas, liver, stomach, kidney, and urinary bladder),19 HPV-related cancers (tongue, mouth, pharynx, rectum/anus, and female genital sites including vagina and vulva),1 and cancers at heavily irradiated sites (small intestine, colon, rectum/anus, urinary bladder, uterine corpus, ovary/fallopian tubes, female genital sites including vagina and vulva, bone, and soft tissue).13 For each second cancer site, as well as for the grouped categories of smoking-related, HPV-related, and heavily irradiated second cancers, we compared SIRs between cervical SCC and AC survivors using Poisson regression. Thus, our analyses included three different comparisons: cervical SCC survivors versus women in the general population (SIR for SCC survivors), cervical AC survivors versus women in the general population (SIR for AC survivors), and comparisons of SIRs for cervical SCC survivors versus cervical AC survivors. All statistical tests were two sided and P values less than .05 were considered as statistically significant.
To further investigate lung cancer risk among cervical SCC and AC survivors, we calculated SIRs separately for second primary SCCs of the lung and second primary ACs of the lung. Expected numbers for each histologic subtype of lung cancer were calculated by applying female general population cancer incidence rates, specific to lung cancer histology, registry, age, calendar year, and race (only for SEER data), to person-years accrued by cervical SCC and AC survivors. These SIRs were calculated overall and after stratification by time since cervical cancer diagnosis (1 to 4, 5 to 9, 10 to 19, and ≥ 20 years).
RESULTS
Characteristics of cervical SCC and AC survivors are presented in Table 1. Age at cervical cancer diagnosis was similar for SCC and AC patients (mean age = 50.2 and 50.0 years, respectively). The proportion of cases of AC histology increased during the study period, with ACs constituting approximately 19% of cervical cancers in the most recent calendar period (1990 and beyond). SCC patients were treated by radiation more frequently than AC patients (51.3% v 41.1%). In the SEER registries, a higher proportion of AC patients than SCC patients were white (82.1% v 75.1%).
Table 1.
Characteristics of Women With Cervical Cancer Stratified by Histology
Characteristic | Condition |
|||
---|---|---|---|---|
Squamous Cell Carcinoma of Cervix |
Adenocarcinoma of Cervix |
|||
No. | % | No. | % | |
No. of patients | 85,109 | 10,280 | ||
No. of person-years | 1,061,885 | 98,704 | ||
Registry | ||||
Denmark (1943-1998) | 24,942 | 29.3 | 2,123 | 20.6 |
US SEER program (1973-2002) | 20,534 | 24.1 | 3,389 | 33.0 |
Sweden (1958-2001) | 19,717 | 23.2 | 2,467 | 24.0 |
Norway (1953-1999) | 12,283 | 14.4 | 1,189 | 11.6 |
Finland (1953-2001) | 7,633 | 9.0 | 1,112 | 10.8 |
Age at cervical cancer diagnosis, years | ||||
< 40 | 22,215 | 26.1 | 2,798 | 27.2 |
40-49 | 21,758 | 25.6 | 2,745 | 26.7 |
50-59 | 17,984 | 21.1 | 2,002 | 19.5 |
60+ | 23,152 | 27.2 | 2,735 | 26.6 |
Calendar year of cervical cancer diagnosis | ||||
< 1960 | 11,724 | 13.8 | 689 | 6.7 |
1960-1974 | 27,768 | 32.6 | 1,860 | 18.1 |
1975-1989 | 28,236 | 33.2 | 3,684 | 35.8 |
1990+ | 17,381 | 20.4 | 4,047 | 39.4 |
Stage of cervical cancer | ||||
Local | 28,356 | 33.3 | 4,153 | 40.4 |
Regional | 9,392 | 11.0 | 1,105 | 10.7 |
Distant | 2,265 | 2.7 | 386 | 3.8 |
Unknown | 45,096 | 53.0 | 4,636 | 45.1 |
Initial treatment | ||||
Any radiotherapy | 43,683 | 51.3 | 4,226 | 41.1 |
No radiotherapy | 20,239 | 23.8 | 3,389 | 32.9 |
Other/unknown | 21,187 | 24.9 | 2,665 | 26.0 |
Race* | ||||
White | 15,421 | 75.1 | 2,783 | 82.1 |
Black | 3,214 | 15.6 | 255 | 7.5 |
Other/unknown | 1,899 | 9.3 | 351 | 10.4 |
Abbreviation: SEER, Surveillance, Epidemiology, and End Results.
Data on race were available only from the US SEER program.
Compared with women in the general population, overall cancer risk was significantly increased among both cervical SCC survivors (Table 2; n = 10,559 second cancers during 1,061,885 person-years of follow-up; SIR, 1.31; 95% CI, 1.29 to 1.34) and cervical AC survivors (n = 920 second cancers during 98,704 person-years of follow-up; SIR, 1.29; 95% CI, 1.22 to 1.38). We also assessed overall second cancer risk separately across calendar period of cervical cancer diagnosis and for SEER and non-SEER registries (Table 2). For both cervical SCC and AC survivors, second cancer risk did not vary significantly by calendar period of cervical cancer diagnosis. In contrast, the overall risk of second cancers was significantly higher among survivors in the non-SEER registries than those in SEER registries (Table 2). Nonetheless, regardless of calendar period of cervical cancer diagnosis or registry, no significant differences were observed in overall second cancer risk between cervical SCC and AC survivors.
Table 2.
Risk of Second Primary Cancers Among Cervical Squamous Cell Carcinoma and Adenocarcinoma Survivors Stratified by Calendar Year of Cervical Cancer Diagnosis and Registry
Parameter | Squamous Cell Carcinoma of Cervix |
Adenocarcinoma of Cervix |
P for Squamous Cell Carcinoma v Adenocarcinoma* | ||
---|---|---|---|---|---|
SIR | 95% CI | SIR | 95% CI | ||
No. of patients | 85,109 | 10,280 | |||
No. of person-years | 1,061,885 | 98,704 | |||
All second cancers except cervix | 1.31 | 1.29 to 1.34 | 1.29 | 1.22 to 1.38 | .592 |
Year of cervical cancer diagnosis | |||||
< 1960 | 1.28 | 1.23 to 1.34 | 1.42 | 1.19 to 1.70 | .244 |
1960-1974 | 1.31 | 1.28 to 1.35 | 1.27 | 1.13 to 1.43 | .560 |
1975-1989 | 1.32 | 1.28 to 1.37 | 1.29 | 1.16 to 1.43 | .594 |
1990+ | 1.40 | 1.30 to 1.51 | 1.26 | 1.06 to 1.50 | .283 |
P for trend | .141† | .361† | |||
Registry | |||||
SEER | 1.20 | 1.15 to 1.26 | 1.16 | 1.02 to 1.33 | .526 |
Non-SEER | 1.34 | 1.31 to 1.37 | 1.34 | 1.25 to 1.45 | .934 |
P | < .001‡ | .035‡ |
Abbreviations: SIR, standardized incidence ratio; SEER, Surveillance, Epidemiology, and End Results.
P values comparing SIR for cervical squamous cell carcinoma v cervical adenocarcinoma survivors were calculated using Poisson regression.
P values for trend in SIRs across calendar periods of cervical cancer diagnosis were calculated using Poisson regression.
P values comparing SIR for SEER versus non-SEER registries were calculated using Poisson regression.
Among both SCC and AC survivors, SIRs for the grouped categories of smoking-related cancers, HPV-related cancers, and cancers at heavily-irradiated sites were significantly increased (Table 3). The most frequent second cancers among both SCC and AC survivors were those of the lung, breast, and colon (Table 3). Among cervical SCC survivors, SIRs for most second cancers were significantly increased. Highest SIRs following cervical SCC were observed for cancers of female genital sites (SIR, 5.00; including vagina, vulva, and other/unspecified genital sites), urinary bladder (SIR, 3.48), and lung (SIR, 2.69). Among cervical AC survivors, significantly increased SIRs were observed for second cancers of the colon, rectum/anus, pancreas, trachea/bronchus/lung, bone, soft tissue, female genital sites, urinary bladder, and non-Hodgkin's lymphoma (NHL). Highest SIRs following cervical AC were observed for cancers of the soft tissue (SIR, 4.70), bone (SIR, 4.66), and female genital sites (SIR, 3.87). Both cervical SCC and AC survivors were at significantly decreased risk of cancers of the breast and uterine corpus than women in the general population (Table 3).
Table 3.
Risk of Second Primary Cancers Among Cervical Cancer Survivors Stratified By Histology of Cervical Cancer
Second Cancer Site | Squamous Cell Carcinoma of Cervix |
Adenocarcinoma of Cervix |
P for Squamous Cell Carcinoma v Adenocarcinoma* | ||||
---|---|---|---|---|---|---|---|
No. of Observed Second Cancers | SIR | 95% CI | No. of Observed Second Cancers | SIR | 95% CI | ||
No. of patients | 85,109 | 10,280 | |||||
No. of person-years | 1,061,885 | 98,704 | |||||
Smoking-related cancers† | 3,923 | 2.07 | 2.01 to 2.14 | 287 | 1.78 | 1.58 to 2.00 | .015 |
HPV-related cancers‡ | 1,248 | 2.30 | 2.18 to 2.43 | 91 | 2.01 | 1.63 to 2.48 | .238 |
Heavily irradiated sites§ | 3,858 | 1.50 | 1.46 to 1.56 | 349 | 1.63 | 1.46 to 1.81 | .242 |
Lip | 26 | 1.66 | 1.09 to 2.44 | 3 | 2.23 | 0.45 to 6.59 | .584 |
Tongue | 28 | 1.25 | 0.83 to 1.81 | 0 | 0.00 | 0.00 to 1.74 | — |
Mouth | 60 | 1.61 | 1.24 to 2.08 | 3 | 0.91 | 0.18 to 2.67 | .333 |
Pharynx | 49 | 2.06 | 1.53 to 2.73 | 0 | 0.00 | 0.00 to 1.77 | — |
Esophagus | 89 | 1.50 | 1.21 to 1.86 | 4 | 0.82 | 0.22 to 2.12 | .239 |
Stomach | 422 | 1.32 | 1.21 to 1.46 | 28 | 1.16 | 0.78 to 1.69 | .527 |
Small intestine | 54 | 1.82 | 1.37 to 2.39 | 3 | 1.17 | 0.24 to 3.44 | .454 |
Colon | 963 | 1.20 | 1.13 to 1.28 | 102 | 1.54 | 1.26 to 1.88 | .017 |
Rectum/anus | 677 | 1.81 | 1.68 to 1.96 | 60 | 1.97 | 1.51 to 2.54 | .523 |
Liver | 73 | 1.13 | 0.89 to 1.43 | 3 | 0.56 | 0.11 to 1.67 | .241 |
Pancreas | 381 | 1.38 | 1.25 to 1.53 | 34 | 1.53 | 1.06 to 2.14 | .551 |
Larynx | 48 | 2.10 | 1.55 to 2.79 | 1 | 0.49 | 0.01 to 2.73 | .149 |
Trachea/bronchus/lung | 1,649 | 2.69 | 2.57 to 2.83 | 126 | 2.18 | 1.82 to 2.61 | .026 |
Bone | 26 | 2.63 | 1.72 to 3.87 | 4 | 4.66 | 1.27 to 12.05 | .289 |
Soft tissue | 89 | 2.36 | 1.90 to 2.91 | 16 | 4.70 | 2.69 to 7.64 | .011 |
Breast | 1,643 | 0.76 | 0.73 to 0.80 | 177 | 0.85 | 0.74 to 1.00 | .142 |
Female genital‖ | 434 | 5.00 | 4.55 to 5.50 | 28 | 3.87 | 2.57 to 5.60 | .194 |
Uterine corpus | 421 | 0.77 | 0.70 to 0.85 | 32 | 0.66 | 0.46 to 0.94 | .271 |
Ovary/fallopian tubes | 383 | 0.87 | 0.79 to 0.97 | 42 | 1.15 | 0.83 to 1.56 | .165 |
Kidney | 287 | 1.34 | 1.20 to 1.51 | 23 | 1.32 | 0.84 to 1.98 | .928 |
Urinary bladder | 811 | 3.48 | 3.25 to 3.74 | 62 | 3.23 | 2.48 to 4.15 | .568 |
Melanoma | 156 | 0.70 | 0.60 to 0.82 | 30 | 1.37 | 0.93 to 1.96 | .001 |
Non-Hodgkin's lymphoma | 272 | 1.25 | 1.11 to 1.41 | 38 | 1.84 | 1.30 to 2.53 | .025 |
Abbreviation: SIR, standardized incidence ratio.
P values comparing SIR for cervical squamous cell carcinoma versus cervical adenocarcinoma survivors were calculated using Poisson regression.
Smoking-related cancers include cancers of the lip, tongue, mouth, pharynx, esophagus, larynx, trachea/bronchus/lung, pancreas, liver, stomach, kidney, and urinary bladder (19).
HPV-related cancers include cancers of the tongue, mouth, pharynx, rectum/anus, and female genital sites including vagina and vulva (1).
Cancers at heavily-irradiated sites include cancers of the small intestine, colon, rectum/anus, urinary bladder, uterine corpus, ovary/fallopian tubes, female genital sites including vagina and vulva, bone, and soft tissue (13).
Female genital cancers include cancers of the vagina, vulva, and other/unspecified genital sites.
We evaluated whether the SIRs for second cancers were significantly different between cervical SCC and AC survivors (P values in Table 3). No significant differences between SCC and AC survivors were observed in the SIRs for the grouped categories of HPV-related cancers or cancers at heavily irradiated sites. In contrast, the SIR for smoking-related second cancers (grouped together) was significantly higher among SCC than AC survivors (SIR, 2.07 v 1.78; P = .015). In particular, risk of second lung cancer was significantly higher among SCC survivors than AC survivors (SIR, 2.69 v 2.18; P = .026). SIRs for other smoking-related second cancers (lip, tongue, mouth, pharynx, esophagus, larynx, pancreas, liver, stomach, kidney, and urinary bladder) were not significantly different between SCC and AC survivors. In addition, SIRs for colon cancer, soft tissue cancers, melanoma, and NHL were significantly higher among AC than SCC survivors (all P < .05).
Risk of second lung cancers was significantly increased among cervical SCC and AC survivors in both the SEER and non-SEER registries (SIRs for SCC = 2.53 and 2.76 and AC = 1.82 and 2.44, respectively). To further evaluate the increased lung cancer risk, we conducted analyses for the major histologic subtypes of lung cancer (Table 4). The 1,775 second primary lung cancers included 666 lung SCCs, 333 lung ACs, and 776 lung cancers of other/unknown histologies. Among cervical SCC survivors, risks of SCC lung cancer (SIR, 5.45) and AC lung cancer (SIR, 1.58) were both significantly elevated. The SIR for SCC lung cancer was exceptionally high during the initial years after cervical cancer diagnosis (SIR during 1 to 4 years, 10.86) and decreased thereafter. The increased risks for lung SCC and lung AC among cervical SCC survivors were evident beyond 20 years after cervical cancer diagnosis. Cervical AC survivors were also at significantly increased risk of both lung SCC and lung AC, and this increased risk was evident up to 20 years after cervical cancer diagnosis. Among both cervical SCC and AC survivors, SIRs for lung SCC were higher than SIRs for lung AC. In addition, the SIR for second lung SCC was higher among cervical SCC than AC survivors, whereas the SIR for second lung AC was higher among cervical AC than SCC survivors.
Table 4.
Risk of Second Lung Cancers Stratified by Histology of Cervical Cancer and Lung Cancer
Parameter | SCC of Cervix |
AC of Cervix |
||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
SCC of Lung |
AC of Lung |
SCC of Lung |
AC of Lung |
|||||||||
Obs | SIR | 95% CI | Obs | SIR | 95% CI | Obs | SIR | 95% CI | Obs | SIR | 95% CI | |
Overall | 635 | 5.45 | 5.04 to 5.90 | 296 | 1.58 | 1.41 to 1.78 | 31 | 3.04 | 2.07 to 4.32 | 37 | 1.96 | 1.38 to 2.71 |
Results according to time after cervical cancer diagnosis, years | ||||||||||||
1-4 | 202 | 10.86 | 9.42 to 12.47 | 55 | 1.79 | 1.35 to 2.33 | 11 | 4.86 | 2.43 to 8.71 | 8 | 1.83 | 0.79 to 3.61 |
5-9 | 202 | 9.88 | 8.57 to 11.35 | 63 | 1.86 | 1.43 to 2.38 | 11 | 5.13 | 2.56 to 9.20 | 10 | 2.36 | 1.13 to 4.36 |
10-19 | 134 | 3.57 | 2.99 to 4.23 | 91 | 1.48 | 1.20 to 1.82 | 8 | 2.44 | 1.05 to 4.82 | 14 | 2.26 | 1.24 to 3.81 |
20+ | 97 | 2.43 | 1.98 to 2.98 | 87 | 1.42 | 1.14 to 1.76 | 1 | 0.39 | 0.01 to 2.23 | 5 | 1.23 | 0.40 to 2.89 |
Abbreviations: SCC, squamous cell carcinoma; AC, adenocarcinoma; Obs, observed number of cases; SIR, standardized incidence ratio.
DISCUSSION
Our study is the first, to our knowledge, to assess second cancer risk separately for cervical SCC and AC survivors. The profiles of second cancers among cervical SCC and AC survivors largely mirror the similarities and differences in cofactors for these two histologic subtypes.3–7 We found that compared with women in the general population, both cervical SCC and AC survivors were at a similar increased risk of all second cancers considered together, HPV-related cancers, and cancers at heavily irradiated sites. In contrast, although significantly increased in both groups when compared with the general population, risk of smoking-related second cancers (particularly lung cancer) was significantly higher among cervical SCC than AC survivors. Cervical SCC and AC survivors also differed in their risk of second cancers of the colon and soft tissue, melanoma, and NHL, with higher risks observed among AC survivors.
Consistent with HPV being a necessary cause of both cervical SCC and AC and with a large proportion of cervical cancer patients receiving radiotherapy, we found that the risks of HPV-related and radiation-related second cancers were similar between cervical SCC and AC survivors. SCC and AC survivors were also similar in their risks for hormone-related second cancers, with significantly decreased risks of cancers of the breast and uterine corpus when compared with women in the general population. These deficits for breast and uterine corpus cancers could be related to treatments for cervical cancer, including hysterectomy and radiation treatment (resulting in ovarian ablation), which reduce subsequent risk of these cancers.17,20
An increased risk of lung cancer among cervical cancer survivors has been demonstrated previously.13–17 However, prior studies did not distinguish cervical cancers by histology, and largely attributed the increased risk to cigarette smoking.13–17 Of note, previous case-control studies and pooled analyses have shown that cigarette smoking is a risk factor for cervical SCC, whereas smoking does not increase risk of cervical AC.3–7 The results of these studies imply that smoking is more common among SCC patients than women in the general population, while the prevalence of smoking among AC patients is similar to that in the general population. Consistent with this formulation, we found that SCC survivors, when compared with women in the general population, were at a significantly increased risk of most smoking-related second cancers. Likewise, the risks of smoking-related second cancers (grouped together) as well as for second lung cancers were significantly higher among cervical SCC than AC survivors (Table 3).
Contrary to expectation, however, we found that AC survivors were also at significantly increased risk of smoking-related second cancers, particularly lung cancer, compared with women in the general population. Given that smoking is not believed to be a cofactor for cervical AC,3–7 this increased lung cancer risk among cervical AC survivors suggests a role for factors other than cigarette smoking. One such factor may be HPV infection, which has been implicated in lung cancer etiology.21,22 Several previous case-series have detected HPV DNA in lung SCC and lung AC tumor tissues.21,22 However, results have been inconsistent across studies, with HPV DNA detection rates in case-series ranging from 0% to 100%.21 A few studies have also reported detection of integrated HPV genomes as well as expression of HPV E6 and E7 oncogenes in lung tumors,23–25 which are characteristic features of cancers caused by HPV.26
An etiologic role for HPV infection in lung cancer may be plausible. Recurrent respiratory papillomatosis (RRP), which is caused by HPV, provides an interesting disease model. 26–28 RRP is characterized by laryngeal papillomas, but can also manifest with tracheal and bronchial papillomas.26–28 Transmission of HPV to the respiratory tract is believed to occur at birth among childhood RRP cases and through oral sex for RRP cases that occur among young adults.27,28 RRP lesions are generally benign, but malignant transformation of papillomas is known to occur.26–28 Likewise, oral/respiratory infection acquired through high-risk sexual behavior, and subsequent malignant transformation, may explain, at least in part, the increased risk of lung cancer after cervical AC that we observed. Alternative explanations for the increased lung cancer risk among cervical AC survivors may include factors such as low SES and high parity, which have been associated with lung cancer risk in the general population.29,30
We considered the possibility that pulmonary metastases, misclassified as primary lung cancers, contributed to the increased risk of lung cancer among cervical cancer survivors. The lung is a common site for distant metastases among cervical cancer patients.31–33 Furthermore, the rate of pulmonary metastases may be higher after cervical AC than cervical SCC.31,33 Indeed, our observations that among cervix SCC survivors, lung SCC risk was exceptionally high in the initial years after cervical cancer, and there was histologic clustering of cervical cancer and lung cancer (ie, risk of lung SCC was higher after cervical SCC than cervical AC, risk of lung AC was higher after cervical AC than after cervical SCC), both suggest that misclassified metastases could have contributed to part of the increased lung cancer risk among cervical cancer survivors. Nevertheless, the increased risk for lung SCC after cervical AC up to 20 years after diagnosis of cervical cancer argues against misclassified metastases as the entire explanation for the increased lung cancer risk among cervical AC survivors.
The highest SIR after cervical SCC was observed for second cancers of female genital sites (including vagina/vulva), whereas highest risk after cervical AC was observed for soft tissue cancers. While the risk of female genital second cancers was similar between cervical SCC and AC survivors, the SIR for soft tissue cancers was significantly higher among cervical AC than SCC survivors. The elevated risk of second cancers at female genital sites among cervical SCC survivors may have arisen from increased exposure to HPV infections and/or high-dose radiation.13,34 However, the reasons for significantly higher risk of soft tissue cancers among cervical AC than SCC survivors are currently unclear. In addition, SIRs for colon cancer, melanoma, and NHL were significantly higher among cervical AC than SCC survivors. These differences between cervical SCC and AC survivors could be related to the relatively higher SES of cervical AC compared with SCC patients. In the general population, colon cancer, melanoma, and NHL are more common among higher SES individuals.35 Although both cervical SCC and AC are associated with low SES,12 cervical AC patients have been shown to be of relatively higher SES than SCC patients.10 One reason for this SES difference may be that Pap smear screening for cervical cancer is more efficient in diagnosing precursors of cervical SCC than of cervical AC.4,5 Thus, in countries with organized cervical cancer screening programs, women diagnosed with invasive cervical SCC tend to be from lower SES groups who are less likely to participate in screening programs.5
We acknowledge several limitations of our study. We did not have information on cofactors for cervical SCC and AC, such as history of smoking, parity, or SES. In contrast to the decreasing incidence of cervical SCC, the incidence of cervical AC has increased during the past few decades in developed countries.2,36 Therefore, although SIRs incorporate an adjustment for calendar year, it is possible that direct comparisons of SIRs between cervical SCC and AC may still be confounded by temporal shifts in the general population cancer patterns. In addition, some of the results we observed may have arisen from misclassification of cervical cancer histology. For the non-SEER registries, we could not exclude cervical adenosquamous cancers, which constitute 20% to 30% of ACs. Nonetheless, it is unclear whether etiologic cofactors differ for AC and adenosquamous cancers,4 and we found that the second cancer profiles in the SEER data were similar irrespective of the inclusion of adenosquamous cancers with ACs (data not shown). We note that increased second cancer risk among cervical cancer survivors, irrespective of histology, may arise from increased medical surveillance among cancer survivors than among women in the general population.37 Finally, it is possible that SIRs for some rare cancers among AC survivors may not have achieved statistical significance owing to low power.
In conclusion, we characterized second cancer risk among cervical AC survivors and compared the patterns of second cancers with cervical SCC survivors. Our results indicate that previously reported similarities and differences in cofactors between cervical SCC and AC translate into similarities and differences in second cancer risk profiles. Notably, although increased in both groups, risks of smoking-related second cancers were significantly higher among cervical SCC than AC survivors. Given that lung cancers constitute a high proportion of all second cancers (approximately 15%) among cervical cancer survivors, further studies are needed to clarify the potential etiologic role of HPV. Our results underscore the importance of cervical cancer cofactors in increasing second cancer risk among survivors.
Acknowledgment
We thank Jeremy Miller, Information Management Services, Rockville, MD, for expert computer support and data management.
Appendix
The Appendix is included in the full-text version of this article, available online at www.jco.org. It is not included in the PDF version (via Adobe® Reader®).
Table A1.
Histologic Classification of Cervical Cancers
Country | Coding System | SCC of Cervix | AC of Cervix |
---|---|---|---|
Denmark | Modified ICD-7 | 171.2 | 171.1 |
US SEER | ICD-O-3 | 8050-8089 | 8140-8231, 8261-8510 |
Sweden | Modified ACS 1951 | 146 | 096 |
Finland | Modified ACS 1951 | 14 | 09 |
Norway | SNOP (tdrough December 31, 1992) | 805x-808x | 814x-838x |
8433 | 8483-8488 | 8493 | 8513 |
8543 | 8573 | 9113 | |
Norway | ICD-O-2 (after January 1, 1993) | 8051 | 7140 |
8052 | 7141 | 8070 | 7142 |
8071 | 7143 | 8072 | 7170 |
8073 | 7508 | 8074 | 8140 |
8075 | 8144 | 8076 | 8200 |
8082 | 8211 | 8260 | 8310 |
8380 | 8430 | 8441 | 8460 |
8480 | 8481 | 8490 | 8510 |
8542 | 8570 | 9110 |
Abbreviations: SCC, squamous cell carcinoma; AC, adenocarcinoma; SEER, Surveillance, Epidemiology, and End Results program; ICD, International Classification of Diseases; ICD-O, International Classification of Diseases for Oncology; ACS, American Cancer Society; SNOP, systematized nomenclature of pathology.
Footnotes
Supported by the Intramural Research Program, National Cancer Institute, National Institutes of Health.
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
The author(s) indicated no potential conflicts of interest.
AUTHOR CONTRIBUTIONS
Conception and design: Anil K. Chaturvedi, Ruth A. Kleinerman, Allan Hildesheim, Ethel S. Gilbert, Eero Pukkala, Lois B. Travis, Eric A. Engels
Financial support: Lois B. Travis, Eric A. Engels
Administrative support: Anil K. Chaturvedi, Lois B. Travis, Eric A. Engels
Provision of study materials or patients: Ruth A. Kleinerman, Hans Storm, Charles F. Lynch, Per Hall, Froydis Langmark, Eero Pukkala, Magnus Kaijser, Michael Andersson, Heikki Joensuu, Lois B. Travis
Collection and assembly of data: Anil K. Chaturvedi, Ruth A. Kleinerman, Hans Storm, Charles F. Lynch, Eero Pukkala, Lois B. Travis
Data analysis and interpretation: Anil K. Chaturvedi, Ruth A. Kleinerman, Allan Hildesheim, Ethel S. Gilbert, Hans Storm, Eero Pukkala, Sophie D. Fossa, Heikki Joensuu, Lois B. Travis, Eric A. Engels
Manuscript writing: Anil K. Chaturvedi, Ruth A. Kleinerman, Allan Hildesheim, Ethel S. Gilbert, Charles F. Lynch, Per Hall, Eero Pukkala, Magnus Kaijser, Sophie D. Fossa, Lois B. Travis, Eric A. Engels
Final approval of manuscript: Anil K. Chaturvedi, Ruth A. Kleinerman, Allan Hildesheim, Ethel S. Gilbert, Hans Storm, Charles F. Lynch, Per Hall, Froydis Langmark, Eero Pukkala, Magnus Kaijser, Michael Andersson, Sophie D. Fossa, Heikki Joensuu, Lois B. Travis, Eric A. Engels
References
- 1.Parkin DM, Bray F. Chapter 2: The burden of HPV-related cancers. Vaccine. 2006;24(suppl 3):S11–S25. doi: 10.1016/j.vaccine.2006.05.111. [DOI] [PubMed] [Google Scholar]
- 2.Wang SS, Sherman ME, Hildesheim A, et al. Cervical adenocarcinoma and squamous cell carcinoma incidence trends among white women and black women in the United States for 1976-2000. Cancer. 2004;100:1035–1044. doi: 10.1002/cncr.20064. [DOI] [PubMed] [Google Scholar]
- 3.Berrington dG, Sweetland S, Green J. Comparison of risk factors for squamous cell and adenocarcinomas of the cervix: A meta-analysis. Br J Cancer. 2004;90:1787–1791. doi: 10.1038/sj.bjc.6601764. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Castellsague X, Diaz M, de Sanjose S, et al. Worldwide human papillomavirus etiology of cervical adenocarcinoma and its cofactors: Implications for screening and prevention. J Natl Cancer Inst. 2006;98:303–315. doi: 10.1093/jnci/djj067. [DOI] [PubMed] [Google Scholar]
- 5.Hildesheim A, Berrington dG. Etiology and prevention of cervical adenocarcinomas. J Natl Cancer Inst. 2006;98:292–293. doi: 10.1093/jnci/djj098. [DOI] [PubMed] [Google Scholar]
- 6.Appleby P, Beral V, Berrington dG, et al. Carcinoma of the cervix and tobacco smoking: Collaborative reanalysis of individual data on 13,541 women with carcinoma of the cervix and 23,017 women without carcinoma of the cervix from 23 epidemiological studies. Int J Cancer. 2006;118:1481–1495. doi: 10.1002/ijc.21493. [DOI] [PubMed] [Google Scholar]
- 7.Lacey JV, Jr, Frisch M, Brinton LA, et al. Associations between smoking and adenocarcinomas and squamous cell carcinomas of the uterine cervix (United States) Cancer Causes Control. 2001;12:153–161. doi: 10.1023/a:1008918310055. [DOI] [PubMed] [Google Scholar]
- 8.Munoz N, Franceschi S, Bosetti C, et al. Role of parity and human papillomavirus in cervical cancer: The IARC multicentric case-control study. Lancet. 2002;359:1093–1101. doi: 10.1016/S0140-6736(02)08151-5. [DOI] [PubMed] [Google Scholar]
- 9.Lacey JV, Jr, Swanson CA, Brinton LA, et al. Obesity as a potential risk factor for adenocarcinomas and squamous cell carcinomas of the uterine cervix. Cancer. 2003;98:814–821. doi: 10.1002/cncr.11567. [DOI] [PubMed] [Google Scholar]
- 10.Brinton LA, Herrero R, Reeves WC, et al. Risk factors for cervical cancer by histology. Gynecol Oncol. 1993;51:301–306. doi: 10.1006/gyno.1993.1294. [DOI] [PubMed] [Google Scholar]
- 11.Green J, Berrington dG, Sweetland S, et al. Risk factors for adenocarcinoma and squamous cell carcinoma of the cervix in women aged 20-44 years: The UK National Case-Control Study of Cervical Cancer. Br J Cancer. 2003;89:2078–2086. doi: 10.1038/sj.bjc.6601296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Parikh S, Brennan P, Boffetta P. Meta-analysis of social inequality and the risk of cervical cancer. Int J Cancer. 2003;105:687–691. doi: 10.1002/ijc.11141. [DOI] [PubMed] [Google Scholar]
- 13.Chaturvedi AK, Engels EA, Gilbert ES, et al. Second cancers among 104,760 survivors of cervical cancer: Evaluation of long-term risk. J Natl Cancer Inst. 2007;99:1634–1643. doi: 10.1093/jnci/djm201. [DOI] [PubMed] [Google Scholar]
- 14.Storm HH. Second primary cancer after treatment for cervical cancer: Late effects after radiotherapy. Cancer. 1988;61:679–688. doi: 10.1002/1097-0142(19880215)61:4<679::aid-cncr2820610411>3.0.co;2-s. [DOI] [PubMed] [Google Scholar]
- 15.Kleinerman RA, Kosary C, Hildesheim A. New malignancies following cancer of the cervix uteri, vagina, and vulva. In: Curtis RE, Freedman DM, Ron E, et al., editors. New Malignancies Among Cancer Survivors: SEER Cancer Registries, 1973-2000. Bethesda, MD: National Cancer Institute; 2006. [Google Scholar]
- 16.Boice JD, Jr, Day NE, Andersen A, et al. Second cancers following radiation treatment for cervical cancer: An international collaboration among cancer registries. J Natl Cancer Inst. 1985;74:955–975. [PubMed] [Google Scholar]
- 17.Kleinerman RA, Boice JD, Jr, Storm HH, et al. Second primary cancer after treatment for cervical cancer: An international cancer registries study. Cancer. 1995;76:442–452. doi: 10.1002/1097-0142(19950801)76:3<442::aid-cncr2820760315>3.0.co;2-l. [DOI] [PubMed] [Google Scholar]
- 18.Liddell FD. Simple exact analysis of the standardised mortality ratio. J Epidemiol Community Health. 1984;38:85–88. doi: 10.1136/jech.38.1.85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Sasco AJ, Secretan MB, Straif K. Tobacco smoking and cancer: A brief review of recent epidemiological evidence. Lung Cancer. 2004;45(suppl 2):S3–S9. doi: 10.1016/j.lungcan.2004.07.998. [DOI] [PubMed] [Google Scholar]
- 20.Boice JD, Jr, Blettner M, Kleinerman RA, et al. Radiation dose and breast cancer risk in patients treated for cancer of the cervix. Int J Cancer. 1989;44:7–16. doi: 10.1002/ijc.2910440103. [DOI] [PubMed] [Google Scholar]
- 21.Syrjanen KJ. HPV infections and lung cancer. J Clin Pathol. 2002;55:885–891. doi: 10.1136/jcp.55.12.885. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Chen YC, Chen JH, Richard K, et al. Lung adenocarcinoma and human papillomavirus infection. Cancer. 2004;101:1428–1436. doi: 10.1002/cncr.20538. [DOI] [PubMed] [Google Scholar]
- 23.Giuliani L, Favalli C, Syrjanen K, et al. Human papillomavirus infections in lung cancer: Detection of E6 and E7 transcripts and review of the literature. Anticancer Res. 2007;27:2697–2704. [PubMed] [Google Scholar]
- 24.Aguayo F, Castillo A, Koriyama C, et al. Human papillomavirus-16 is integrated in lung carcinomas: A study in Chile. Br J Cancer. 2007;97:85–91. doi: 10.1038/sj.bjc.6603848. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Cheng YW, Wu MF, Wang J, et al. Human papillomavirus 16/18 E6 oncoprotein is expressed in lung cancer and related with p53 inactivation. Cancer Res. 2007;67:10686–10693. doi: 10.1158/0008-5472.CAN-07-1461. [DOI] [PubMed] [Google Scholar]
- 26.Gillison ML, Shah KV. Chapter 9: Role of mucosal human papillomavirus in nongenital cancers. J Natl Cancer Inst Monogr. 2003:57–65. doi: 10.1093/oxfordjournals.jncimonographs.a003484. [DOI] [PubMed] [Google Scholar]
- 27.Lacey CJ, Lowndes CM, Shah KV. Chapter 4: Burden and management of non-cancerous HPV-related conditions: HPV-6/11 disease. Vaccine. 2006;24(suppl 3):S35–S41. doi: 10.1016/j.vaccine.2006.06.015. [DOI] [PubMed] [Google Scholar]
- 28.Kashima HK, Shah F, Lyles A, et al. A comparison of risk factors in juvenile-onset and adult-onset recurrent respiratory papillomatosis. Laryngoscope. 1992;102:9–13. doi: 10.1288/00005537-199201000-00002. [DOI] [PubMed] [Google Scholar]
- 29.Ekberg-Aronsson M, Nilsson PM, Nilsson JA, et al. Socio-economic status and lung cancer risk including histologic subtyping: A longitudinal study. Lung Cancer. 2006;51:21–29. doi: 10.1016/j.lungcan.2005.08.014. [DOI] [PubMed] [Google Scholar]
- 30.Kabat GC, Miller AB, Rohan TE. Reproductive and hormonal factors and risk of lung cancer in women: A prospective cohort study. Int J Cancer. 2007;120:2214–2220. doi: 10.1002/ijc.22543. [DOI] [PubMed] [Google Scholar]
- 31.Drescher CW, Hopkins MP, Roberts JA. Comparison of the pattern of metastatic spread of squamous cell cancer and adenocarcinoma of the uterine cervix. Gynecol Oncol. 1989;33:340–343. doi: 10.1016/0090-8258(89)90524-6. [DOI] [PubMed] [Google Scholar]
- 32.Shin MS, Shingleton HM, Partridge EE, et al. Squamous cell carcinoma of the uterine cervix: Patterns of thoracic metastases. Invest Radiol. 1995;30:724–729. doi: 10.1097/00004424-199512000-00006. [DOI] [PubMed] [Google Scholar]
- 33.Imachi M, Tsukamoto N, Matsuyama T, et al. Pulmonary metastasis from carcinoma of the uterine cervix. Gynecol Oncol. 1989;33:189–192. doi: 10.1016/0090-8258(89)90549-0. [DOI] [PubMed] [Google Scholar]
- 34.Insinga RP, Liaw KL, Johnson LG, et al. A systematic review of the prevalence and attribution of human papillomavirus types among cervical, vaginal, and vulvar precancers and cancers in the United States. Cancer Epidemiol Biomarkers Prev. 2008;17:1611–1622. doi: 10.1158/1055-9965.EPI-07-2922. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Schottenfeld D, Fraumeni JF Jr, editors. Cancer Epidemiology and Prevention. ed 3. New York, NY: Oxford University Press; 2006. [Google Scholar]
- 36.Bray F, Carstensen B, Moller H, et al. Incidence trends of adenocarcinoma of the cervix in 13 European countries. Cancer Epidemiol Biomarkers Prev. 2005;14:2191–2199. doi: 10.1158/1055-9965.EPI-05-0231. [DOI] [PubMed] [Google Scholar]
- 37.Boice JD., Jr . Ionizing radiation. In: Schottenfeld D, Fraumeni JF Jr, editors. Cancer Epidemiology and Prevention. ed 3. New York, NY: Oxford University Press; 2006. pp. 259–293. [Google Scholar]