Lack of Significant Effects of Chlamydia trachomatis Infection on Cervical Adenocarcinoma Risk: Nested Case-Control Study.

BACKGROUND: A role of Chlamydia trachomatis in HPV-induced cervical carcinogenesis has been reported for cervical cancer but studies on cervical adenocarcinoma are limited.
METHODS: A total of 1,553 cervical smears taken up to 26 years before diagnosis in a large population-based nested case-control study of cervical adenocarcinoma (AC, 132 cases with matched controls), and adenocarcinoma in situ (AIS, 159 cases with matched controls) were tested for C. trachomatis and HPV DNA by a type-specific PCR bead-based multiplex genotyping (TS-MPG) assay.
RESULTS: Only 1.7% of samples were positive for C. trachomatis, with no significant differences between AC/AIS cases and controls. HPV-positivity was detected in 49.3% of C. trachomatis-negative and 65.4% C. trachomatis-positive samples, respectively.
CONCLUSIONS: A large prospective study did not find any risk for cervical adenocarcinoma and/or AIS conferred by C. trachomatis infection. IMPACT: C. trachomatis appears not to be involved in cervical adenocarcinomas.

Introduction

Persistent infection with human papillomavirus (HPV) is a virtually necessary cause of cervical cancer [1]. One of the most common curable sexually transmitted infections (STI) worldwide Chlamydia trachomatis has been associated with an increased risk of cervical cancer [2-4]. While the most of the studies are focused on squamous cervical carcinoma, there was more limited power to study cervical adenocarcinoma, but no association was found [3,5], which is surprising as C. trachomatis primarily infects glandular cells [2]. Some studies relied on seroepidemiology, where it is difficult to rule out residual confounding completely [1]. One study reported no detection of C. trachomatis DNA in 71 archived formalin-fixed paraffin-embedded tissues of cervical adenocarcinoma (5), but cross-sectional studies are less informative than longitudinal studies [1] and C. trachomatis has only been detected in samples taken many years before the cancer, not close to the diagnosis [2].

We previously reported a large prospective study of cervical adenocarcinoma that established a strong association with HPV present many years before cancer diagnosis (6). We now wished to use this large prospective study to investigate possible associations between C. trachomatis and cervical adenocarcinoma.

Materials and Methods

Detailed characteristics of the study have been reported previously [6]. A total of 1,553 β-globin-positive cervical smears collected during the pre-HPV vaccine era (1969–2002) in Sweden before the development of adenocarcinoma in-situ (AIS, 133 cases; mean age at entry and diagnosis: 28 and 37 years, respectively) or invasive cervical adenocarcinoma (AC, 170 cases; 37/43 years) and matched controls (128 and 169; 29/37 and 37/43 years, respectively) were tested for the presence of C. trachomatis and 19 mucosal HPV types (HPV-6, 11, 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, 68a and 68b, 70, 73 and 82), using a validated type-specific PCR bead-based multiplex genotyping (TS-MPG, IARC, Lyon, France) assay that combines multiplex polymerase chain reaction (PCR) and bead-based Luminex technology (Luminex Corp., Austin, TX, USA), as described elsewhere [7,8].

Results

In total, 1.7% (26/1553) of samples were positive for C. trachomatis (Table 1). The 26 samples corresponded to only 21 distinct subjects. No major difference in C. trachomatis-positivity was observed between the outcomes and case-control status; the odds-ratios (ORs) for the association between chlamydia infection in the first collected smear and subsequent adenocarcinoma were 1.25 for AIS (95% CI 0.34–4.65) and 3.0 for AC (95%CI 0.31–28.84); ORs associated with chlamydia infection in the last smear before diagnosis were not estimable due to the lack of exposed cases (AIS) and controls (AC), respectively (Table 2). The full data set is available as Supplementary Information (S1 Table).

Table 1

Study population and distribution of C. trachomatis positive smears between cases and controls, for all β-globin-positive smears collected in the study.

	Adenocarcinoma in situ (AIS)		Invasive adenocarcinoma (AC)
	Cases	Controls	Cases	Controls
Subjects	133	128	170	169
Age at first smeara	28 (17–71)	29 (16–71)	37 (17–82)	37 (17–82)
Age at last smeara	33 (17–72)	34 (17–71)	41 (20–82)	39 (20–82)
Age at diagnosisa	37 (20–75)	37 (20–75)	43 (25–88)	43 (25–89)
Time in studya	7.4 (0.1–27.0)	8.0 (0.2–25.9)	6.1 (0.0–21.1)	6.0 (0.1–22.2)
Smearsb	425 (100)	332 (100)	422 (100)	374 (100)
HPV positiveb	323 (76)	93 (28)	267 (63)	87 (23)
CT positiveb	10 (2.4)	7 (2.1)	7 (1.7)	2 (0.5)

a Reported in years: median (minimum-maximum)

b Reported as count (percentage)

CT–C. trachomatis HPV–Human papillomavirus (any strain)

Table 2

Risk of cervical adenocarcinoma given by C. trachomatis infection in the first and last smear, calculated as odds ratio via conditional logistic regression where possible.

	Smear	Matched cases/controls	Exposed cases/controlsa	Odds ratio	95% Conf.int.
AIS	First	128	5/4	1.25	0.34–4.65
	Last	128	0/2	NAb	NAb
AC	First	158	3/1	3.00	0.31–28.84
	Last	158	2/0	NAb	NAb

a Matched case/control pairs that differ in exposure and contribute to the OR estimation, reported as exposed cases/exposed controls

b Could not be calculated due to perfect separation (no exposed cases or controls)

Footnote: AIS–adenocarcinoma in situ, AC–adenocarcinoma

HPV-positivity was detected in 49.3% (n = 753) of C. trachomatis-negative and 65.4% (n = 17) C. trachomatis-positive samples. Because of the overall low numbers, a further analysis of possible interaction between C. trachomatis and HPV status was not feasible.

Discussion

In the current largest prospective study to date of C. trachomatis and cervical adenocarcinoma, with a follow-up period of up to 26 years, C. trachomatis was not associated with increased risks of subsequent invasive adenocarcinoma (AC) and its precursor, adenocarcinoma in situ (AIS). The average age of participants in this study was above 30 years, and it is possible that co-infections/interactions between C. trachomatis and HPV, or other STI, may be of more relevance at earlier age and this may require further studying. A higher prevalence of HPV infections have been found in younger Italian women affected with a C. trachomatis chronic infection from a STI centre than in the ones from an assisted reproductive technology clinic [9]. HPV genotype distribution showed that mostly uncommon low risk genotypes were associated with C. trachomatis [10]. However, while Danish women who reported more than one Chlamydia infection had a statistically significantly increased risk of CIN3+, no association was found between C. trachomatis DNA and subsequent risk of CIN3+ among the ones who were HPV-positive or had a persistent HPV infection at baseline [11]. No association between C. trachomatis status, as assessed by DNA or IgG, and risk of cervical premalignancy, after controlling for carcinogenic HPV-positive status was found in a previous study from the United States [12], which suggested that positive associations between C. trachomatis and cervical premalignancy could have been caused, in part, by an increased susceptibility to HPV infection [12]. However, no studies have been done among the women with cervical adenocarcinoma and the present study is the first of this kind. Moreover, the reports of an interaction between C. trachomatis and HPV in squamous cell carcinoma of the cervix are mechanistically unexplained [1] and further studies on the co-factor role of genital microbiota in promoting malignancies, in particular in high-risk and younger populations, may be warranted.

The analysis dataset on the detection of Chlamydia trachomatis and human papillomavirus (HPV) infections in women with cervical adenocarcinoma.

SubjectID–unique subject identifier, RisksetID–unique risk set identifier, CaseStatus–case-control status (1 = case), AgeAtDiag–age at the time of diagnosis, HPVpos–HPV-positivity (1 = HPV-positive), CTpos–Chlamydia trachomatis-positivity (1 = Chlamydia-positive), AgeAtSmear–age at the time of collecting cervial smear).

(ODS)

Click here for additional data file.