- Email: [email protected]
Influence of a loop electrosurgical excision procedure (LEEP) on levels of cytokines in cervical secretions
Journal of Reproductive Immunology 109 (2015) 74–83
Contents lists available at ScienceDirect
Journal of Reproductive Immunology journal homepage: www.elsevier.com/locate/jreprimm
Influence of a loop electrosurgical excision procedure (LEEP) on levels of cytokines in cervical secretions Audrey F. Saftlas a,∗ , Cassandra N. Spracklen b , Kelli K. Ryckman c , Colleen K. Stockdale d , Kerri Penrose e , Kevin Ault f , Linda M. Rubenstein g , Ligia A. Pinto h a
Department of Epidemiology, University of Iowa College of Public Health, 145 Riverside Drive, S427 CPHB, Iowa City, IA 52242, USA Department of Genetics, University of North Carolina-Chapel Hill, 5100 Genetic Medicine Building, CB #7264, 120 Mason Farm Road, Chapel Hill, NC 27599, USA c Department of Epidemiology, University of Iowa College of Public Health, 145 Riverside Drive, S435 CPHB, Iowa City, IA 52242, USA d Department of Obstetrics and Gynecology, Carver College of Medicine, University of Iowa, 200 Hawkins Drive, Iowa City, IA 51214, USA e Department of Medicine, Division of Infectious Disease, University of Pittsburgh, S804 Scaife Hall, 3550 Terrace Street, Pittsburgh, PA 15261, USA f Department of Obstetrics and Gynecology, University of Kansas Medical Center, 3901 Rainbow Boulevard, Mailstop 2028, Kansas City, KS 66160, USA g Department of Epidemiology, University of Iowa College of Public Health, 145 Riverside Drive, S415 CPHB, Iowa City, IA 52242, USA h Human Papillomavirus Immunology Laboratory, Leidos Biomedical Research, Incorporated, Frederick National Laboratory for Cancer Research, 1050 Boyles Street, Frederick, MD, USA b
a r t i c l e
i n f o
Article history: Received 8 September 2014 Received in revised form 14 November 2014 Accepted 8 January 2015 Keywords: Loop electrosurgical excision procedure (LEEP) Cervical intraepithelial neoplasia (CIN) Cervical secretions Cytokines
a b s t r a c t Over the past decade, there has been heightened interest in determining if there is an increased risk of adverse reproductive outcomes among women who had a loop electrosurgical excision procedure (LEEP) to remove cervical intraepithelial neoplasia (CIN). The objective of this exploratory study was to determine if the treatment of CIN with a LEEP is associated with changes in cervical soluble immune markers. Cervical cytokine concentrations were measured in women treated with LEEP and a control group of women who had colposcopy only and did not undergo LEEP. Cytokines were examined in cervical secretions collected in Merocel® sponges at study entry and at 6-month follow-up. Cytokines were measured using a Luminex 18-plex cytokine bead assay. The mean cytokine levels were not significantly changed from baseline to follow-up in either group, with the exception of TNF-␣, which decreased among women who underwent a LEEP. When the mean levels of cytokines of the treated and untreated groups at baseline or follow-up were compared, cytokine levels tended to be lower in the treated group (particularly IFN-␥, IL-6, IL-8, and MCP-1). Findings from adjusted repeated measures analyses revealed no differences between the two groups with regard to changes in cytokine levels over time. Overall, women undergoing a LEEP showed few changes in the cervical microenvironment relative to untreated women. Future studies with additional cervical environment markers and larger sample sizes are needed to determine if a LEEP is associated with dysregulation of the cervical microenvironment. © 2015 Elsevier Ireland Ltd. All rights reserved.
∗ Corresponding author. E-mail addresses: [email protected] (A.F. Saftlas), [email protected] (C.N. Spracklen), [email protected] (K.K. Ryckman), [email protected] (C.K. Stockdale), [email protected] (K. Penrose), [email protected] (K. Ault), [email protected] (L.M. Rubenstein), [email protected] (L.A. Pinto). http://dx.doi.org/10.1016/j.jri.2015.01.002 0165-0378/© 2015 Elsevier Ireland Ltd. All rights reserved.
A.F. Saftlas et al. / Journal of Reproductive Immunology 109 (2015) 74–83
1. Introduction Pap smears to detect precancerous cervical intraepithelial neoplasia (CIN) are a routine component of a woman’s health care program. As a result of such screening, women with biopsy-confirmed CIN, particularly CIN2 or CIN3, most often undergo a loop electrosurgical excision procedure (LEEP) to remove cervical dysplasia. Over the past decade, there has been renewed interest in studying the influence of a LEEP on the increased risk of adverse reproductive effects, including preterm delivery (Sadler et al., 2004; Kyrgiou et al., 2006) and subfertility (Spracklen et al., 2013). Although the mechanisms underlying these associations have not been established, the potential for LEEP to have an adverse impact on reproductive success is important, as 500,000 women of reproductive age in the US alone are treated annually (Del Priore et al., 2010). As the mucosal immune environment of the female reproductive tract is responsible for numerous functions, from first-line defense to the facilitation and maintenance of pregnancy (Rodriguez-Garcia et al., 2013), it is possible that the mucosal cytokine microenvironment can be disturbed by the cervical tissue removal that occurs with surgery, which in turn could have an impact on fertility. Cytokine dysregulation resulting from gynecological disorders, such as sexually transmitted infections and pelvic inflammatory disease, have been reported to negatively influence fertility (Iwabe et al., 2002; Jerchel et al., 2014). Inflammatory immune markers can negatively affect ovulation, hormones required for reproduction, sperm and egg quality, and implantation (Weiss et al., 2009). Local levels of cervical cytokines and other immunerelated markers have been investigated in a number of studies in the context of HPV infection, cervical cancer, and other genital infections. Findings are variable, often based on small sample sizes, a small number of markers, and cross-sectional study designs. There is an overall consensus, however, that a Th1-type cytokine response is critical for clearing an HPV infection. In contrast, HPV persistence may be associated with increased levels of chronic inflammation, as reflected by increased levels of pro-inflammatory cytokines (Song et al., 2008; Kemp et al., 2010; Marks et al., 2011; Mhatre et al., 2012; Rosa et al., 2012; Scott et al., 2013; Iwata et al., 2014; Paradkar et al., 2014), which may in turn lead to increased levels of immunosuppressive cytokines, Th2type cytokines/chemokines, and decreased recruitment of Th1 effector responses, all of which are hallmarks of progression to cancer (Kemp et al., 2010; Marks et al., 2011; Mhatre et al., 2012; Rosa et al., 2012; Scott et al., 2013; Iwata et al., 2014). To our knowledge, no studies have investigated the effect of a LEEP on the soluble mucosal cytokine microenvironment. To explore the extent of changes in the cervical cytokine microenvironment following a LEEP, we designed a longitudinal study to compare levels of 18 different cytokines, inflammatory markers, and chemokines. We hypothesized that the cytokine changes linked to tissue healing and remodeling following tissue removal with LEEP would alter the reproductive
75
microenvironment, which in turn could impair reproductive success. 2. Materials and methods 2.1. Study design and subject enrollment This prospective study recruited two groups of women who underwent colposcopy at the University of Iowa between September 2007 and October 2010. Women who required a LEEP following colposcopy were recruited into the LEEP group (n = 65) and those who did not were recruited into the untreated group (n = 76). Cervical secretion samples were collected at the time of colposcopy for untreated women and just before the LEEP for treated women. All subjects returned to the clinic six months later for follow-up and provided a post-healing cervical secretion sample. The follow-up period of six months was chosen because cervical tissue healing is complete within six months post-procedure (Paraskevaidis et al., 2002); additionally, the clinic schedules six-month followup appointments for LEEP patients. Eligible women were 18–38 years of age with regular menstrual cycles; were not pregnant and had not been pregnant in the past six months; had no previous cervical surgery, dilatation and curettage procedure, or abortive outcomes; and had no history of CIN, cancer, HIV/AIDS or autoimmune disease. Written informed consent was obtained before collecting the cervical secretion samples and the demographic and clinical data. All protocols and informed consent procedures were approved by the University of Iowa Institutional Review Board. 2.2. Sample collection Before sampling at baseline and follow-up, subjects underwent a routine pelvic examination for clinical signs of infection (including vaginal/cervical discharge and high vaginal pH), were tested for pregnancy, and were screened for chlamydia and gonorrhea. Wet mount studies were conducted to detect clue cells and trichomonas. Cervical secretions were collected by a single practitioner using Merocel® (XOMED Surgical Products, Jacksonville, FL, USA) ophthalmic sponges using sampling methods previously described (Hildesheim et al., 1999). Briefly, the sponge portion of the ophthalmic device was placed gently into the cervical os and kept in place for 30 s. Upon removal, the sponge was immediately placed into a cryovial and into liquid nitrogen and then transferred to a −70 ◦ C freezer. Samples were not collected from women who were currently menstruating; who had used oral steroids within the past 2 weeks, inhaled steroids within the past 24 h, or emergency contraception in the past 30 days; who had douched or engaged in vaginal intercourse within the past 48 h; who had a vaginal or sexually transmitted infection; or who became pregnant. 2.3. Laboratory analysis All specimens were processed and analyzed at the Frederick National Laboratory for Cancer Research (FNLCR;
76
A.F. Saftlas et al. / Journal of Reproductive Immunology 109 (2015) 74–83
Frederick, MD, USA). Cervical secretions were extracted from the sponges using a previously established protocol (Koshiol et al., 2014). First, the sponges were weighed and recorded. Then, 300 l of extraction buffer (PBS; Invitrogen, Grand Island, NY, USA), 256 mM NaCl (Sigma–Aldrich, St. Louis, MO, USA), and 100 g/ml aprotinin (Sigma–Aldrich) were slowly added to the sponge and incubated at 4 ◦ C for 30 min. The lysate was centrifuged (13,000 × g for 15 min at 4 ◦ C). This process was repeated once, and then 4 l of fetal bovine serum was added to the extract. The extracts were aliquoted and frozen at −80 ◦ C until further testing. We applied a dilution factor to account for variation in the amount of cervical secretions collected from each woman. The dilution factor was calculated as [(x − y) + 0.6 g of buffer]/(x − y), where x equals the weight of the sponge after collection, and y is the weight of the dry sponge (Koshiol et al., 2014). The weight of the dry sponge was calculated as the average of the weights of 12 dry, sterile sponges from the same batch as those used in the clinical study. Eighteen cytokines were measured in cervical fluid using a Luminex 18-plex bead array cytokine assay (EGF, G-CSF, GM-CSF, IFN-␥, IL-1␣, IL-1, IL-1RA, IL-2, IL-4, IL6, IL-8, IL-10, IL-12, IL-13, MCP-1, MIP-1␣, RANTES, and TNF-␣; LINCO Research, St. Charles, MO, USA), covering a number of immune functions, such as inflammatory, Th1 and Th2-type, in addition to anti-inflammatory. The two samples from any given woman were run on the same plate, and the two treatment groups were randomly distributed on each plate to balance the sample distribution between different batches and to mitigate the effects of inter-run variability. The samples were tested in duplicate wells and analyzed as per the manufacturer’s instructions using the BioPlex 100 Analyzer (Bio-Rad Laboratories, Hercules, CA, USA); concentrations are expressed in pg/ml. IL-4 and MIP-1␣ had a low percentage of subjects with detectable levels (<50%) and were excluded from subsequent analyses. 2.4. Statistical methods Cytokine values below the limit of detection were assigned a value equivalent to the lowest detectable level. For all statistical analyses, cytokine levels were logtransformed to normalize the distribution of values. All data analyses were conducted using SAS 9.3 (SAS Institute, Cary, NC, USA) and graphs were generated using GraphPad Prism 6. To identify factors associated with a LEEP, we compared the characteristics of participants by treatment group and tested for statistical significance using Chi-squared tests for categorical variables. To determine if a LEEP was associated with changes in cytokine levels, we conducted a series of descriptive analyses. First, we compared and tested for differences in the mean concentration (logtransformed) of each cytokine at baseline and follow-up within and between the two groups using Student’s ttest. Cytokines with significant differences in mean levels within and/or between groups over time were displayed as box plots depicting the inter-quartile range and minimum and maximum values. Categorical analyses were
conducted to determine if the frequency distribution of cytokine levels divided into tertiles differed within and/or between the two groups at baseline and follow-up. Tertiles were based on the distribution of each cytokine within the untreated group at baseline. Subjects with nondetectable levels were included in the lowest tertile. Differences in tertiles between the LEEP and untreated groups at baseline and at six-month follow-up were assessed using global Chi-squared tests. Pearson’s correlation coefficients were estimated to determine the correlation between baseline and follow-up cytokine levels within treatment groups. Differences in the correlation between treatment groups were assessed using Fisher’s r to z transformation. Although these analyses involved multiple comparisons, we did not apply a Bonferroni correction because of the exploratory nature of the study and concern that correcting for multiple comparisons might hide results worth pursuing. Unadjusted and adjusted repeated measures analyses were performed using PROC GLIMMIX to determine if cytokine levels changed significantly over time. In this analysis, the variable, “LEEP group,” tested the null hypothesis of a common intercept between the LEEP and No-LEEP groups. The “Time” variable tested for changes from baseline to follow-up. An interaction term, “LEEP group × Time,” was constructed to determine if changes in cytokines over time differed between treatment groups. Potentially confounding variables examined included: age, smoking status, contraception method, education, race, alcohol consumption, recreational drug use, and number of sexual partners. Variables that changed the point estimates by at least 10% were retained in the final analysis. 3. Results Forty treated (LEEP) and 49 untreated (i.e., colposcopy only) women successfully completed the study; their characteristics by treatment group are shown in Table 1. Women requiring a LEEP were significantly older, had more lifetime sexual partners, and were more likely to smoke than untreated women, but there were no differences in terms of race, type of contraception used, douching habits, frequency of sexual intercourse, and alcohol and recreational drug use. Mean levels for each of 16 cytokines were examined for differences within and between treatment groups over time (Table 2), as were the corresponding tertile distributions of cytokine levels (Supplemental Table 1). Cytokines with mean levels that changed significantly over time (p < 0.10), either within or between groups, are displayed as box plots in Fig. 1 and described below. Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.jri.2015.01.002. 3.1. IFN- At follow-up, mean levels of IFN- were lower in the LEEP versus the No-LEEP group (203.6 versus 316.7 pg/ml, p = 0.06). At baseline, there were no significant differences in the mean levels of IFN- (252.6 and 331.4 pg/ml, in
A.F. Saftlas et al. / Journal of Reproductive Immunology 109 (2015) 74–83
77
Table 1 Characteristics of the 89 study participants by LEEP status in the Cervical Composition Study, Iowa City, IA, 2007–2010. Characteristics Age group 18–20 21–24 25–29 30–38 Race White Asian Unknown Type of contraceptive used None Monophase OC Triphase OC Nuva Ring Depo-Provera Mini Pill Mirena IUD Patch Biphase Nonhormonal IUD OC unknown Frequency of douching Never
LEEP group (n = 40)
No LEEP group (n = 49)
1 (2.5) 15 (37.5) 13 (32.5) 11 (27.5)
8 (16.3) 23 (46.9) 12 (24.5) 6 (12.2)
40 (100.0) 0 (0.0) 0 (0.0)
48 (98.0) 1 (2.0) 0 (0.0)
1 (2.5) 17 (42.5) 11 (27.5) 4 (10.0) 3 (7.5) 2 (5.0) 2 (5.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0)
5 (10.2) 22 (44.9) 12 (24.5) 2 (4.1) 2 (4.1) 2 (4.1) 2 (4.1) 1 (2.0) 1 (2.0) 0 (0.0) 0 (0.0)
36 (90.0) 3 (7.5) 1 (2.5)
49 (100.0) 0 (0.0) 0 (0.0)
36 (90.0) 2 (5.0) 1 (2.5) 1 (2.5)
49 (100.0) 0 (0.0) 0 (0.0) 0 (0.0)
1 (2.5) 13 (32.5) 18 (45.0) 8 (20.0)
4 (8.2) 24 (49.0) 17 (35.7) 4 (8.2)
1 (2.5) 3 (7.5) 11 (27.5) 17 (42.5) 3 (7.5) 5 (12.5)
0 (0.0) 12 (24.5) 21 (42.9) 10 (20.4) 6 (12.2) 0 (0.0)
29 (72.5) 11 (27.5)
41 (83.7) 8 (16.3)
2 (5.0) 4 (10.0) 3 (7.5) 5 (12.5) 12 (30.0) 14 (35.0)
4 (8.2) 6 (12.2) 4 (8.2) 8 (16.3) 12 (24.5) 15 (30.6)
7 (18.4) 5 (13.2) 14 (36.8) 6 (15.8) 6 (15.8)
8 (18.6) 11 (25.6) 17 (39.5) 5 (11.6) 2 (4.7)
38 (95.0) 2 (5.0)
45 (93.8) 3 (6.3)
p valuea
0.03
0.36
0.72
0.08
0.16
0.13
0.01
0.04
0.96
0.35
0.80
Abbreviations: LEEP, loop electrosurgical excision procedure; OC, oral contraceptive; IUD, intrauterine device. Results presented as n (%). a Chi-squared p value for the group difference LEEP versus No LEEP.
the LEEP and No-LEEP groups respectively) between the two groups. Within-group differences from baseline to follow-up were not significant, despite a trend toward a decrease in IFN- among treated women.
3.2. IL-6 At baseline, LEEP subjects had significantly lower mean levels of IL-6 than untreated subjects
78
Table 2 Mean analysis of cytokine levels detected at baseline and six-month follow-up by LEEP status in the Cervical Composition Study, Iowa City, IA, 2007–2010. Cytokine
Follow-up
Mean within group p valuec
Not detectable, n (%)
Mean ± SDa
Range
Not detectable, n (%)
Mean ± SD
7 (17.5) 8 (16.3)
788.1 ± 513.1 882.5 ± 712.6 0.47b
164.9–2694.9 69.3–2909.8
27 (32.5) 6 (12.2)
692.1 ± 466.1 858.0 ± 817.9 0.23
0 (0.0) 0 (0.0)
111,886.0 ± 84,704.2 137,158.5 ± 111,365.0 0.24
9893.0–371,104.6 19,652.4–445,726.0
0 (0.0) 0 (0.0)
200,624.7 ± 466.1 115,595.3 ± 103,328.5 0.19
0 (0.0) 0 (0.0)
124.4 ± 118.8 150.9 ± 106.4 0.27
26.0–518.7 25.3–482.6
0 (0.0) 0 (0.0)
105.8 ± 73.6 142.2 ± 135.5 0.11
2 (5.0) 6 (12.2)
252.6 ± 193.7 331.4 ± 373.0 0.20
4.2–806.5 2.6–1304.6
5 (12.5) 7 (14.3)
203.6 ± 184.8 316.7 ± 368.0 0.06
0 (0.0) 0 (0.0)
6197.0 ± 6086.4 5359.6 ± 4824.3 0.48
305.3–30,729.5 66.4–27,495.6
0 (0.0) 0 (0.0)
13,866.0 ± 20,607.1 11,764.9 ± 32,006.7 0.70
0 (0.0) 0 (0.0)
1354.9 ± 2630.7 1004.9 ± 1578.7 0.44
22.7–12,493.8 17.4–8822.4
0 (0.0) 0 (0.0)
1104.1 ± 2056.8 1244.4 ± 2040.7 0.75
0 (0.0) 0 (0.0)
378,992.8 ± 405,930.3 334,996.5 ± 485,628.2 0.65
28,664.2–1,832,433.1 5955.8–2,908,601.4
0 (0.0) 0 (0.0)
756,088.7 ± 1,122,296.0 479,288.2 ± 574,858.2 0.16
3 (7.5) 2 (4.1)
114.5 ± 92.7 124.1 ± 109.6 0.65
4.2–411.5 12.1–435.5
5 (12.5) 4 (8.2)
0 (0.0) 0 (0.0)
5953.9 ± 5131.8 9353.9 ± 8447.0 0.02
572.0–17,445.5 565.3–40,717.3
0 (0.0) 0 (0.0)
Range 60.1–2059.6 64.2–3388.8
0.20 0.56
1115.3–2,023,863.6 11,488.9–409,327.6
0.98 0.19
8.1–368.8 14.6–664.8
0.47 0.24
3.2–710.4 2.6–1548.1
0.12 0.75
109.3–108,827.0 206.4–226,036.3
0.22 0.43
32.1–8810.5 15.9–8546.5
0.98 0.82
59,316.2–5,360,000.7 9856.7–2,875,478.7
0.11 0.17
89.4 ± 77.1 131.6 ± 166.3 0.12
2.4–381.6 7.9–852.3
0.26 0.62
5289.8 ± 5293.6 7085.7 ± 6652.3 0.16
42.8–22,611.0 257.9–29,971.8
0.29 0.11
A.F. Saftlas et al. / Journal of Reproductive Immunology 109 (2015) 74–83
EGF LEEP No-LEEP Between groups p value G-CSF LEEP No-LEEP Between groups p value GM-CSF LEEP No-LEEP Between groups p value IFN-␥ LEEP No-LEEP Between groups p value IL-1␣ LEEP No-LEEP Between groups p value IL-1 LEEP No-LEEP Between groups p value IL-1RA LEEP No-LEEP Between groups p value IL-2 LEEP No-LEEP Between groups p value IL-6 LEEP No-LEEP Between groups p value
Baseline
0 (0.0) 0 (0.0)
85,885.4 ± 83,575.0 174,818.6 ± 210,300.1 0.01
4523.9–380,668.9 3646.4–1,159,634.3
0 (0.0) 0 (0.0)
99,534.3 ± 91,033.1 133,154.9 ± 143,326.4 0.18
1 (2.5) 0 (0.0)
753.8 ± 1020.5 901.1 ± 1346.3 0.56
6.6–4656.7 28.0–7833.5
0 (0.0) 0 (0.0)
469.6 ± 577.4 1141.1 ± 3528.7 0.19
1 (2.5) 1 (2.0)
208.0 ± 207.4 214.4 ± 213.1 0.89
4.2–1173.7 8.2–979.7
3 (7.5) 2 (4.1)
154.3 ± 120.6 213.4 ± 217.5 0.11
5.0–603.5 2.6–895.6
0.24 0.85
7 (17.5) 2 (4.1)
89.2 ± 91.6 95.9 ± 83.8 0.72
3.5–387.9 12.0–403.2
7 (17.5) 4 (8.2)
63.9 ± 60.5 96.5 ± 105.4 0.07
5.0–371.5 2.6–560.9
0.43 0.57
0 (0.0) 0 (0.0)
18,230.8 ± 127,252.3 21,218.7 ± 27,398.0 0.61
751.9–118,963.1 1813.8–144,560.7
0 (0.0) 0 (0.0)
8929.1 ± 8912.3 15,642.6 ± 20,516.1 0.04
301.0–36,327.4 354.0–110,058.6
0.11 0.08
0 (0.0) 0 (0.0)
4636.8 ± 16,336.8 5035.7 ± 11,645.2 0.89
33.0–102,791.1 70.7–74,380.0
0 (0.0) 0 (0.0)
3941.8 ± 18,972.5 13,826.8 ± 50,411.9 0.21
30.2–120,433.3 48.4–290,726.7
0.30 0.94
4 (10.0) 3 (6.1)
588.8 ± 1124.7 391.1 ± 490.9 0.31
6.6–6203.0 4.0–2170.5
9 (22.5) 5 (10.2)
267.3 ± 527.2 329.5 ± 817.3 0.67
2293.4–352,857.3 7172.6–586,302.1
7.5–2956.4 11.4–24,859.8
2.4–2581.1 2.6–5695.1
0.57 0.46
0.25 0.63
0.03 0.20
A.F. Saftlas et al. / Journal of Reproductive Immunology 109 (2015) 74–83
IL-8 LEEP No-LEEP Between groups p value IL-10 LEEP No-LEEP Between groups p value IL-12 LEEP No-LEEP Between groups p value IL-13 LEEP No-LEEP Between groups p value MCP-1 LEEP No-LEEP Between groups p value RANTES LEEP No-LEEP Between groups p value TNF-␣ LEEP No-LEEP Between groups p value
LEEP, n = 40; No-LEEP, n = 49. Abbreviations: LEEP, loop electrosurgical excision procedure; SD, standard deviation; IQR, inter-quartile range. a Cytokines are presented in units of pg/ml. b t-test comparing the means between the LEEP and No-LEEP groups at baseline and at follow-up using the log-transformed cytokine values. c t-test comparing the means between baseline and follow-up cytokine values within the LEEP and No-LEEP groups using the log-transformed cytokine values.
79
80
A.F. Saftlas et al. / Journal of Reproductive Immunology 109 (2015) 74–83
Fig. 1. Box plots of the mean analyses of cytokine levels detected at baseline and six-month follow-up by LEEP status in the Cervical Composition Study, Iowa City, IA 2007–2010. Whiskers represent the minimum and maximum values detected for each cytokine. “+” represents the mean value. “*” represents statistically significant differences between the means of p < 0.10. “**” represents statistically significant differences between the means of p < 0.05. Abbreviations: BL, baseline; FU, follow-up.
(5953.9–9353.9 pg/ml, p = 0.02). Consistent with these findings, there were differences in the tertile distribution of cytokine levels at baseline between the two treatment groups (p = 0.04; Supplemental Table 1).
174,816.6 pg/ml, p = 0.01). There were also differences in the tertile distribution of IL-8 at baseline between the treatment groups, with lower levels in LEEP subjects (Supplemental Table 1).
3.3. IL-8
3.4. IL-13
As with IL-6, LEEP subjects at baseline had significantly lower levels of IL-8 than untreated women (85,885.4 versus
At baseline, the two groups had similar mean levels of IL-13. Mean levels of IL-13 at follow-up, however,
A.F. Saftlas et al. / Journal of Reproductive Immunology 109 (2015) 74–83
were lower in the LEEP group (63.9 versus 96.5 pg/ml, p = 0.07).
Table 3 Summary of correlations from the log-transformed baseline × follow-up cytokine plots.
3.5. MCP-1 As observed with IL-13 and IFN-, the mean level of MCP-1 at follow-up was significantly lower in untreated women (8929.1 versus 15,642.6 pg/ml, p = 0.04). The mean level of MCP-1 among untreated women was also lower at follow-up than at baseline (15,642.6 versus 21,218.7 pg/ml, p = 0.08). 3.6. TNF-˛ Among LEEP subjects, the mean level of TNF-˛ was significantly lower at follow-up than at baseline (588.8–267.3 pg/ml, p = 0.03); in contrast, the comparable mean levels in untreated women were unchanged (391.1–329.5 pg/ml, p = 0.20). As shown in Table 3, correlations of log-transformed cytokine levels at baseline and follow-up in untreated women were significant for all except RANTES; among LEEP subjects, significant correlations were observed for all but five cytokines: IL-1ˇ, IL-8, IL-13, MCP-1, and RANTES. Baseline and follow-up cytokine levels, however, were more highly correlated in the untreated versus treated women for seven cytokines: EGF, G-CSF, IL-2, IL-10, IL-12, IL-13, and TNF-˛. We conducted repeated measures analyses to determine if cytokine levels changed significantly over time (Supplemental Table 2). Adjusted results indicate that the LEEP and No-LEEP groups shared a common intercept for all cytokines except for IL-8 (p = 0.07) and RANTES (p = 0.03). The slopes of the LEEP and No-LEEP groups were not significantly different for any of the 16 cytokines, indicating that changes in cytokine levels over time between the two groups were not different. Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.jri.2015.01.002. 4. Discussion This study is the first to investigate the effect of LEEP on the soluble mucosal cytokine microenvironment. We evaluated changes in levels of 16 cervical cytokines in treated and untreated subjects using a well-qualified multiplex technology and assays (Chaturvedi et al., 2011; Koshiol et al., 2014) for paired longitudinal assessment. Contrary to our hypothesis, changes in cytokine levels over time of the two groups were not significantly different. In addition, LEEP did not significantly affect levels of cytokines in our samples, with the exception of TNF-˛, which decreased significantly among treated women over the study period. Overall, these findings suggest that LEEP did not substantially affect levels of the selected cytokines and chemokines with key immunological functions in the cervical cytokine microenvironment. Chronic inflammation, characterized by increased levels of TNF-˛ in cervical fluid has been associated with persistent HPV infection (Scott et al., 2013) and cervical disease
81
Pearson’s correlation
p valuea
p valueb
EGF
LEEP No-LEEP
0.47 0.74
0.002 <0.0001
0.04
G-CSF
LEEP No-LEEP
0.37 0.66
0.02 <0.0001
0.07
GM-CSF
LEEP No-LEEP
0.48 0.63
0.002 <0.0001
0.32
IFN-␥
LEEP No-LEEP
0.31 0.52
0.04 0.0001
0.25
IL-10
LEEP No-LEEP
0.36 0.71
0.02 <0.0001
0.02
IL-12
LEEP No-LEEP
0.35 0.71
0.03 <0.0001
0.02
IL-13
LEEP No-LEEP
0.18 0.69
0.28 <0.0001
0.003
IL-1␣
LEEP No-LEEP
0.43 0.51
0.005 0.0002
0.64
IL-1
LEEP No-LEEP
0.18 0.46
0.27 0.001
0.15
IL-1RA
LEEP No-LEEP
0.33 0.49
0.04 0.0004
0.38
IL-2
LEEP No-LEEP
0.32 0.63
0.04 <0.0001
0.06
IL-6
LEEP No-LEEP
0.35 0.49
0.02 0.0003
0.44
IL-8
LEEP No-LEEP
0.06 0.35
0.71 0.01
0.17
MCP-1
LEEP No-LEEP
0.12 0.36
0.45 0.01
0.25
RANTES
LEEP No-LEEP
0.17 0.16
0.31 0.28
0.96
TNF-␣
LEEP No-LEEP
0.42 0.71
0.008 <0.0001
0.04
a
p value for the Pearson’s correlation of the log-transformed cytokine values and LEEP or No-LEEP group. b p value comparing the correlations for each log-transformed cytokine value between the LEEP and No-LEEP groups using Fisher’s r-to-z transformation.
progression. We found that baseline levels of TNF-˛ tended to be higher in women who required LEEP, though this difference was not statistically significant. Following LEEP, levels of TNF-˛ decreased significantly to levels that were comparable with those among untreated subjects. Levels of TNF-˛ did not change over time in untreated subjects. One possible explanation for the decrease in the mean level of TNF-˛ in the LEEP group is that the treatment eliminated the lesions persistently infected with HPV, which may have been the source or inducer of increased baseline levels of TNF-˛ in the cervical secretions of treated women. The biological and clinical implications of these findings are unknown. Because of the small sample size, assessment of potential interactive effects between levels of TNF-˛ and other cytokines was not feasible. An unexpected observation was that mean cytokine levels in treated women, all of whom had a diagnosis of CIN2
82
A.F. Saftlas et al. / Journal of Reproductive Immunology 109 (2015) 74–83
or CIN3, tended to be lower than those among untreated women, who had CIN1 or lower. Mean cytokine levels of IFN-˛, IL-6, IL-8, IL-13, and MCP-1 were lower at baseline and/or follow-up in treated women. These results are consistent with reports of decreased levels of IFN-˛ and IL-6 in lesions of advanced cervical disease (Song et al., 2008; Rosa et al., 2012), but disagree with others where higher levels of IL-6 and IL-8 were found in cervical lavage samples from women with CIN versus those without CIN (Mhatre et al., 2012). Several environmental, genetic, and constitutional factors can cause levels of cytokines to vary widely from woman to woman, even among women of similar ages and phases of the menstrual cycle (Franklin and Kutteh, 1999). While such factors may explain the wide range of cytokine levels observed in our study, we applied strict eligibility criteria to narrow the window of cervical secretion sampling and minimize the influence of menstrual cycle timing, hormonal contraceptive use, intercourse, and genital infections. Furthermore, fluctuations in cytokine levels in paired longitudinal measurements of cervical secretions are not surprising given that the cervix is an extremely complex and dynamic environment composed of a large number of cell types that produce a wide array of cytokines, chemokines, growth factors, and anti-microbial peptides captured in cervical secretions (Rodriguez-Garcia et al., 2013). Different sampling methods have been used to measure immune markers (cervical or cervico-vaginal swabs, sponges, and lavages). Multiplexed, bead-based assays to measure immune/inflammation-related markers in cervical secretions have been widely used as a tool to measure cytokine profiles in serum and other fluids in the context of cancer and infections. We have successfully used the multiplex magnetic bead assays to identify pre-diagnostic immune/inflammatory markers of risk for a number of cancers (Purdue et al., 2013; Shiels et al., 2013); and more recently, we optimized and validated the use of Luminex bead arrays in cervical secretions (Koshiol et al., 2014). Currently, there are no standardized methods or procedures for the collection and measurement of cytokines in cervical secretions, making it difficult to compare our findings with those of other studies. We performed adjustments to our measures based on the weight of each sample to normalize for differences in volume collection between different women. Some studies using cervical lavages or sponges have not applied normalizations, making comparisons of studies a very challenging task. In addition, different assay platforms and kit manufacturers have been used in different studies, which likely have contributed to differences in results and an inability to directly compare the absolute levels of cytokines of different studies. Our study is rigorous in several other respects. All cervical secretion samples were collected among women from a single clinic, by a single practitioner, using the same clinical method and protocol, and analyzed using kits from the same batch of a single manufacturer. All study subjects recruited had colposcopy; therefore, untreated women were likely to be similar to treated women on factors that we could not adjust for analytically. We also applied a stringent set of exclusion criteria over the study period to reduce
the likelihood that other factors known to influence levels of cervical cytokines were in play. This study is also subject to some limitations. As pathology reports and HPV DNA information were unavailable, it was not possible to examine for a direct association of cytokine levels with HPV infection and cervical disease. During the study period, however, the colposcopy service at the University of Iowa uniformly treated women with a diagnosis of CIN2 or higher and did not perform a LEEP in women with CIN1 or lower. Another potential limitation could be the relatively brief follow-up period of six months. While tissue healing is complete after six months, it is possible that relevant immunological changes secondary to a LEEP may not be apparent within this relatively brief time period. Additional limitations include the relatively small sample size and the chance that some statistically significant findings are an artifact of multiple comparisons. To address some of these limitations, we are planning a larger study to investigate more directly the potential associations between cytokine alterations, persistent infection, and associated cervical disease. 5. Conclusion This exploratory study of 16 cytokines suggests that levels of TNF-˛ were significantly reduced following a LEEP, although levels of the other cytokines were not significantly changed. Because of the relatively small sample size and limited set of cytokines evaluated, further research is needed to determine if LEEP leads to dysregulation of other elements of the cervical cytokine microenvironment. Standardization of procedures and methods for collecting cervical secretions and analyzing the cervical cytokine milieu is urgently needed to allow for fair comparisons between different laboratories. Conflict of interest statement None declared. Acknowledgements This work was supported by the National Institute of Allergy and Infectious Diseases (R21-AI06811). This project has been funded in whole or in part by federal funds from the National Cancer Institute, National Institutes of Health, under Contract no. HHSN261200800001E. The content of this paper does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does the mention of trade names, commercial products, or organizations imply endorsement by the US Government. We thank Dr. Phillip Castle for contributing to the conceptualization of the project, and Mary Cherrico for coordinating and conducting the clinical aspects of this study. References Chaturvedi, A.K., Kemp, T.J., Pfeiffer, R.M., Biancotto, A., Williams, M., Munuo, S., et al., 2011. Evaluation of multiplexed cytokine and
A.F. Saftlas et al. / Journal of Reproductive Immunology 109 (2015) 74–83 inflammation marker measurements: a methodologic study. Cancer Epidemiol. Biomarkers Prev. 20, 1902–1911. Del Priore, G., Gudipudi, D.K., Montemarano, N., Restivo, A.M., Malanowska-Stega, J., Arslan, A.A., 2010. Oral diindolylmethane (DIM): pilot evaluation of a nonsurgical treatment for cervical dysplasia. Gynecol. Oncol. 116, 464–467. Franklin, R.D., Kutteh, W.H., 1999. Characterization of immunoglobulins and cytokines in human cervical mucus: influence of exogenous and endogenous hormones. J. Reprod. Immunol. 42, 93–106. Hildesheim, A., McShane, L.M., Schiffman, M., Bratti, M.C., Rodriguez, A.C., Herrero, R., et al., 1999. Cytokine and immunoglobulin concentrations in cervical secretions: reproducibility of the Weck-cel collection instrument and correlates of immune measures. J. Immunol. Methods 225, 131–143. Iwabe, T., Harada, T., Terakawa, N., 2002. Role of cytokines in endometriosis-associated infertility. Gynecol. Obstet. Invest. 53 (Suppl. 1), 19–25. Iwata, T., Fujii, T., Morii, K., Saito, M., Sugiyama, J., Nishio, H., et al., 2014. Cytokine profile in cervical mucosa of Japanese patients with cervical intraepithelial neoplasia. Int. J. Clin. Oncol., http://dx.doi.org/10. 1007/s10147-014-0680-8. Jerchel, S., Kaufhold, I., Shuchardt, L., Shima, K., Rupp, J., 2014. Host immune responses after hypoxic reactivation of IFN-gamma induced persistent chlamydia trachomatis infection. Front. Cell. Infect. Microbiol. 4, 43. Kemp, T.J., Hildesheim, A., Garcia-Pineres, A., Williams, M.C., Shearer, G.M., Rodriguez, A.C., et al., 2010. Elevated systemic levels of inflammatory cytokines in older women with persistent cervical human papillomavirus infection. Cancer Epidemiol. Biomarkers Prev. 19, 1954–1959. Koshiol, J., Sklavos, M., Wentzensen, N., Kemp, T., Schiffman, M., Dunn, S.T., et al., 2014. Evaluation of a multiplex panel of immune-related markers in cervical secretions: a methodologic study. Int. J. Cancer 134, 411–425. Kyrgiou, M., Koliopoulos, G., Martin-Hirsch, P., Arbyn, M., Prendiville, W., Paraskevaidis, E., 2006. Obstetric outcomes after conservative treatment for intraepithelial or early invasive cervical lesions: systematic review and meta-analysis. Lancet 367, 489–498. Marks, M.A., Viscidi, R.P., Chang, K., Silver, M., Burke, A., Howard, R., et al., 2011. Differences in the concentration and correlation of cervical immune markers among HPV positive and negative perimenopausal women. Cytokine 56, 798–803.
83
Mhatre, M., McAndrew, T., Carpenter, C., Burk, R.D., Einstein, M.H., Herold, B.C., 2012. Cervical intraepithelial neoplasia is associated with genital tract mucosal inflammation. Sex. Transm. Dis. 39, 591–597. Paradkar, P.H., Joshi, J.V., Mertia, P.N., Agashe, S.V., Vaidya, R.A., 2014. Role of cytokines in genesis, progression and prognosis of cervical cancer. Asian Pac. J. Cancer Prev. 15, 3851–3864. Paraskevaidis, E., Bilirakis, E., Koliopoulos, G., Lolis, E.D., Kalantaridou, S., Paschopoulos, M., et al., 2002. Cervical regeneration after diathermy excision of cervical intraepithelial neoplasia as assessed by transvaginal sonography. Eur. J. Obstet. Gynecol. Reprod. Biol. 102, 88–91. Purdue, M.P., Hofmann, J.N., Kemp, T.J., Chaturvedi, A.K., Lan, Q., Park, J.H., et al., 2013. A prospective study of 67 serum immune and inflammation markers and risk of non-Hodgkin lymphoma. Blood 122, 951–957. Rodriguez-Garcia, M., Patel, M.V., Wira, C.R., 2013. Innate and adaptive anti-HIV immune responses in the female reproductive tract. J. Reprod. Immunol. 97, 74–84. Rosa, M.I., Moraes, M.V., Vuolo, F., Petronilho, F., Bozzetti, M.C., Medeiros, L.R., et al., 2012. Association of interleukin-6 in women with persistence of DNA-HPV: a nested case-control study. Arch. Gynecol. Obstet. 285, 143–148. Sadler, L., Saftlas, A.F., Wang, W., Exeter, M., Whittaker, J., McCowan, L., 2004. Treatment for cervical intraepithelial neoplasia and risk of preterm delivery. JAMA 291, 2100–2106. Scott, M.E., Shvetsov, Y.B., Thompson, P.J., Hernandez, B.Y., Zhu, X., Wilkens, L.R., et al., 2013. Cervical cytokines and clearance of incident human papillomavirus infection: Hawaii HPV cohort study. Int. J. Cancer 133, 1187–1196. Shiels, M.S., Pfeiffer, R.M., Hildersheim, A., Engels, E.A., Kemp, T.J., Park, J.H., et al., 2013. Circulating inflammation markers and prospective risk for lung cancer. J. Natl. Cancer Inst. 105, 1871–1880. Song, S.H., Lee, J.K., Lee, N.W., Saw, H.S., Kang, J.S., Lee, K.W., 2008. Interferon-gamma (IFN-gamma): a possible prognostic marker for clearance of high-risk human papillomavirus (HPV). Gynecol. Oncol. 108, 543–548. Spracklen, C.N., Harland, K.K., Stegmann, B.J., Saftlas, A.F., 2013. Cervical surgery for cervical intraepithelial neoplasia and prolonged time to conception of a live birth: a case-control study. Br. J. Obstet. Gynaecol. 120, 960–965. Weiss, G., Goldsmith, L.T., Taylor, R.N., Bellet, D., Taylor, H.S., 2009. Inflammation in reproductive disorders. Reprod. Sci. 16, 216–229.
Contents lists available at ScienceDirect
Journal of Reproductive Immunology journal homepage: www.elsevier.com/locate/jreprimm
Influence of a loop electrosurgical excision procedure (LEEP) on levels of cytokines in cervical secretions Audrey F. Saftlas a,∗ , Cassandra N. Spracklen b , Kelli K. Ryckman c , Colleen K. Stockdale d , Kerri Penrose e , Kevin Ault f , Linda M. Rubenstein g , Ligia A. Pinto h a
Department of Epidemiology, University of Iowa College of Public Health, 145 Riverside Drive, S427 CPHB, Iowa City, IA 52242, USA Department of Genetics, University of North Carolina-Chapel Hill, 5100 Genetic Medicine Building, CB #7264, 120 Mason Farm Road, Chapel Hill, NC 27599, USA c Department of Epidemiology, University of Iowa College of Public Health, 145 Riverside Drive, S435 CPHB, Iowa City, IA 52242, USA d Department of Obstetrics and Gynecology, Carver College of Medicine, University of Iowa, 200 Hawkins Drive, Iowa City, IA 51214, USA e Department of Medicine, Division of Infectious Disease, University of Pittsburgh, S804 Scaife Hall, 3550 Terrace Street, Pittsburgh, PA 15261, USA f Department of Obstetrics and Gynecology, University of Kansas Medical Center, 3901 Rainbow Boulevard, Mailstop 2028, Kansas City, KS 66160, USA g Department of Epidemiology, University of Iowa College of Public Health, 145 Riverside Drive, S415 CPHB, Iowa City, IA 52242, USA h Human Papillomavirus Immunology Laboratory, Leidos Biomedical Research, Incorporated, Frederick National Laboratory for Cancer Research, 1050 Boyles Street, Frederick, MD, USA b
a r t i c l e
i n f o
Article history: Received 8 September 2014 Received in revised form 14 November 2014 Accepted 8 January 2015 Keywords: Loop electrosurgical excision procedure (LEEP) Cervical intraepithelial neoplasia (CIN) Cervical secretions Cytokines
a b s t r a c t Over the past decade, there has been heightened interest in determining if there is an increased risk of adverse reproductive outcomes among women who had a loop electrosurgical excision procedure (LEEP) to remove cervical intraepithelial neoplasia (CIN). The objective of this exploratory study was to determine if the treatment of CIN with a LEEP is associated with changes in cervical soluble immune markers. Cervical cytokine concentrations were measured in women treated with LEEP and a control group of women who had colposcopy only and did not undergo LEEP. Cytokines were examined in cervical secretions collected in Merocel® sponges at study entry and at 6-month follow-up. Cytokines were measured using a Luminex 18-plex cytokine bead assay. The mean cytokine levels were not significantly changed from baseline to follow-up in either group, with the exception of TNF-␣, which decreased among women who underwent a LEEP. When the mean levels of cytokines of the treated and untreated groups at baseline or follow-up were compared, cytokine levels tended to be lower in the treated group (particularly IFN-␥, IL-6, IL-8, and MCP-1). Findings from adjusted repeated measures analyses revealed no differences between the two groups with regard to changes in cytokine levels over time. Overall, women undergoing a LEEP showed few changes in the cervical microenvironment relative to untreated women. Future studies with additional cervical environment markers and larger sample sizes are needed to determine if a LEEP is associated with dysregulation of the cervical microenvironment. © 2015 Elsevier Ireland Ltd. All rights reserved.
∗ Corresponding author. E-mail addresses: [email protected] (A.F. Saftlas), [email protected] (C.N. Spracklen), [email protected] (K.K. Ryckman), [email protected] (C.K. Stockdale), [email protected] (K. Penrose), [email protected] (K. Ault), [email protected] (L.M. Rubenstein), [email protected] (L.A. Pinto). http://dx.doi.org/10.1016/j.jri.2015.01.002 0165-0378/© 2015 Elsevier Ireland Ltd. All rights reserved.
A.F. Saftlas et al. / Journal of Reproductive Immunology 109 (2015) 74–83
1. Introduction Pap smears to detect precancerous cervical intraepithelial neoplasia (CIN) are a routine component of a woman’s health care program. As a result of such screening, women with biopsy-confirmed CIN, particularly CIN2 or CIN3, most often undergo a loop electrosurgical excision procedure (LEEP) to remove cervical dysplasia. Over the past decade, there has been renewed interest in studying the influence of a LEEP on the increased risk of adverse reproductive effects, including preterm delivery (Sadler et al., 2004; Kyrgiou et al., 2006) and subfertility (Spracklen et al., 2013). Although the mechanisms underlying these associations have not been established, the potential for LEEP to have an adverse impact on reproductive success is important, as 500,000 women of reproductive age in the US alone are treated annually (Del Priore et al., 2010). As the mucosal immune environment of the female reproductive tract is responsible for numerous functions, from first-line defense to the facilitation and maintenance of pregnancy (Rodriguez-Garcia et al., 2013), it is possible that the mucosal cytokine microenvironment can be disturbed by the cervical tissue removal that occurs with surgery, which in turn could have an impact on fertility. Cytokine dysregulation resulting from gynecological disorders, such as sexually transmitted infections and pelvic inflammatory disease, have been reported to negatively influence fertility (Iwabe et al., 2002; Jerchel et al., 2014). Inflammatory immune markers can negatively affect ovulation, hormones required for reproduction, sperm and egg quality, and implantation (Weiss et al., 2009). Local levels of cervical cytokines and other immunerelated markers have been investigated in a number of studies in the context of HPV infection, cervical cancer, and other genital infections. Findings are variable, often based on small sample sizes, a small number of markers, and cross-sectional study designs. There is an overall consensus, however, that a Th1-type cytokine response is critical for clearing an HPV infection. In contrast, HPV persistence may be associated with increased levels of chronic inflammation, as reflected by increased levels of pro-inflammatory cytokines (Song et al., 2008; Kemp et al., 2010; Marks et al., 2011; Mhatre et al., 2012; Rosa et al., 2012; Scott et al., 2013; Iwata et al., 2014; Paradkar et al., 2014), which may in turn lead to increased levels of immunosuppressive cytokines, Th2type cytokines/chemokines, and decreased recruitment of Th1 effector responses, all of which are hallmarks of progression to cancer (Kemp et al., 2010; Marks et al., 2011; Mhatre et al., 2012; Rosa et al., 2012; Scott et al., 2013; Iwata et al., 2014). To our knowledge, no studies have investigated the effect of a LEEP on the soluble mucosal cytokine microenvironment. To explore the extent of changes in the cervical cytokine microenvironment following a LEEP, we designed a longitudinal study to compare levels of 18 different cytokines, inflammatory markers, and chemokines. We hypothesized that the cytokine changes linked to tissue healing and remodeling following tissue removal with LEEP would alter the reproductive
75
microenvironment, which in turn could impair reproductive success. 2. Materials and methods 2.1. Study design and subject enrollment This prospective study recruited two groups of women who underwent colposcopy at the University of Iowa between September 2007 and October 2010. Women who required a LEEP following colposcopy were recruited into the LEEP group (n = 65) and those who did not were recruited into the untreated group (n = 76). Cervical secretion samples were collected at the time of colposcopy for untreated women and just before the LEEP for treated women. All subjects returned to the clinic six months later for follow-up and provided a post-healing cervical secretion sample. The follow-up period of six months was chosen because cervical tissue healing is complete within six months post-procedure (Paraskevaidis et al., 2002); additionally, the clinic schedules six-month followup appointments for LEEP patients. Eligible women were 18–38 years of age with regular menstrual cycles; were not pregnant and had not been pregnant in the past six months; had no previous cervical surgery, dilatation and curettage procedure, or abortive outcomes; and had no history of CIN, cancer, HIV/AIDS or autoimmune disease. Written informed consent was obtained before collecting the cervical secretion samples and the demographic and clinical data. All protocols and informed consent procedures were approved by the University of Iowa Institutional Review Board. 2.2. Sample collection Before sampling at baseline and follow-up, subjects underwent a routine pelvic examination for clinical signs of infection (including vaginal/cervical discharge and high vaginal pH), were tested for pregnancy, and were screened for chlamydia and gonorrhea. Wet mount studies were conducted to detect clue cells and trichomonas. Cervical secretions were collected by a single practitioner using Merocel® (XOMED Surgical Products, Jacksonville, FL, USA) ophthalmic sponges using sampling methods previously described (Hildesheim et al., 1999). Briefly, the sponge portion of the ophthalmic device was placed gently into the cervical os and kept in place for 30 s. Upon removal, the sponge was immediately placed into a cryovial and into liquid nitrogen and then transferred to a −70 ◦ C freezer. Samples were not collected from women who were currently menstruating; who had used oral steroids within the past 2 weeks, inhaled steroids within the past 24 h, or emergency contraception in the past 30 days; who had douched or engaged in vaginal intercourse within the past 48 h; who had a vaginal or sexually transmitted infection; or who became pregnant. 2.3. Laboratory analysis All specimens were processed and analyzed at the Frederick National Laboratory for Cancer Research (FNLCR;
76
A.F. Saftlas et al. / Journal of Reproductive Immunology 109 (2015) 74–83
Frederick, MD, USA). Cervical secretions were extracted from the sponges using a previously established protocol (Koshiol et al., 2014). First, the sponges were weighed and recorded. Then, 300 l of extraction buffer (PBS; Invitrogen, Grand Island, NY, USA), 256 mM NaCl (Sigma–Aldrich, St. Louis, MO, USA), and 100 g/ml aprotinin (Sigma–Aldrich) were slowly added to the sponge and incubated at 4 ◦ C for 30 min. The lysate was centrifuged (13,000 × g for 15 min at 4 ◦ C). This process was repeated once, and then 4 l of fetal bovine serum was added to the extract. The extracts were aliquoted and frozen at −80 ◦ C until further testing. We applied a dilution factor to account for variation in the amount of cervical secretions collected from each woman. The dilution factor was calculated as [(x − y) + 0.6 g of buffer]/(x − y), where x equals the weight of the sponge after collection, and y is the weight of the dry sponge (Koshiol et al., 2014). The weight of the dry sponge was calculated as the average of the weights of 12 dry, sterile sponges from the same batch as those used in the clinical study. Eighteen cytokines were measured in cervical fluid using a Luminex 18-plex bead array cytokine assay (EGF, G-CSF, GM-CSF, IFN-␥, IL-1␣, IL-1, IL-1RA, IL-2, IL-4, IL6, IL-8, IL-10, IL-12, IL-13, MCP-1, MIP-1␣, RANTES, and TNF-␣; LINCO Research, St. Charles, MO, USA), covering a number of immune functions, such as inflammatory, Th1 and Th2-type, in addition to anti-inflammatory. The two samples from any given woman were run on the same plate, and the two treatment groups were randomly distributed on each plate to balance the sample distribution between different batches and to mitigate the effects of inter-run variability. The samples were tested in duplicate wells and analyzed as per the manufacturer’s instructions using the BioPlex 100 Analyzer (Bio-Rad Laboratories, Hercules, CA, USA); concentrations are expressed in pg/ml. IL-4 and MIP-1␣ had a low percentage of subjects with detectable levels (<50%) and were excluded from subsequent analyses. 2.4. Statistical methods Cytokine values below the limit of detection were assigned a value equivalent to the lowest detectable level. For all statistical analyses, cytokine levels were logtransformed to normalize the distribution of values. All data analyses were conducted using SAS 9.3 (SAS Institute, Cary, NC, USA) and graphs were generated using GraphPad Prism 6. To identify factors associated with a LEEP, we compared the characteristics of participants by treatment group and tested for statistical significance using Chi-squared tests for categorical variables. To determine if a LEEP was associated with changes in cytokine levels, we conducted a series of descriptive analyses. First, we compared and tested for differences in the mean concentration (logtransformed) of each cytokine at baseline and follow-up within and between the two groups using Student’s ttest. Cytokines with significant differences in mean levels within and/or between groups over time were displayed as box plots depicting the inter-quartile range and minimum and maximum values. Categorical analyses were
conducted to determine if the frequency distribution of cytokine levels divided into tertiles differed within and/or between the two groups at baseline and follow-up. Tertiles were based on the distribution of each cytokine within the untreated group at baseline. Subjects with nondetectable levels were included in the lowest tertile. Differences in tertiles between the LEEP and untreated groups at baseline and at six-month follow-up were assessed using global Chi-squared tests. Pearson’s correlation coefficients were estimated to determine the correlation between baseline and follow-up cytokine levels within treatment groups. Differences in the correlation between treatment groups were assessed using Fisher’s r to z transformation. Although these analyses involved multiple comparisons, we did not apply a Bonferroni correction because of the exploratory nature of the study and concern that correcting for multiple comparisons might hide results worth pursuing. Unadjusted and adjusted repeated measures analyses were performed using PROC GLIMMIX to determine if cytokine levels changed significantly over time. In this analysis, the variable, “LEEP group,” tested the null hypothesis of a common intercept between the LEEP and No-LEEP groups. The “Time” variable tested for changes from baseline to follow-up. An interaction term, “LEEP group × Time,” was constructed to determine if changes in cytokines over time differed between treatment groups. Potentially confounding variables examined included: age, smoking status, contraception method, education, race, alcohol consumption, recreational drug use, and number of sexual partners. Variables that changed the point estimates by at least 10% were retained in the final analysis. 3. Results Forty treated (LEEP) and 49 untreated (i.e., colposcopy only) women successfully completed the study; their characteristics by treatment group are shown in Table 1. Women requiring a LEEP were significantly older, had more lifetime sexual partners, and were more likely to smoke than untreated women, but there were no differences in terms of race, type of contraception used, douching habits, frequency of sexual intercourse, and alcohol and recreational drug use. Mean levels for each of 16 cytokines were examined for differences within and between treatment groups over time (Table 2), as were the corresponding tertile distributions of cytokine levels (Supplemental Table 1). Cytokines with mean levels that changed significantly over time (p < 0.10), either within or between groups, are displayed as box plots in Fig. 1 and described below. Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.jri.2015.01.002. 3.1. IFN- At follow-up, mean levels of IFN- were lower in the LEEP versus the No-LEEP group (203.6 versus 316.7 pg/ml, p = 0.06). At baseline, there were no significant differences in the mean levels of IFN- (252.6 and 331.4 pg/ml, in
A.F. Saftlas et al. / Journal of Reproductive Immunology 109 (2015) 74–83
77
Table 1 Characteristics of the 89 study participants by LEEP status in the Cervical Composition Study, Iowa City, IA, 2007–2010. Characteristics Age group 18–20 21–24 25–29 30–38 Race White Asian Unknown Type of contraceptive used None Monophase OC Triphase OC Nuva Ring Depo-Provera Mini Pill Mirena IUD Patch Biphase Nonhormonal IUD OC unknown Frequency of douching Never
LEEP group (n = 40)
No LEEP group (n = 49)
1 (2.5) 15 (37.5) 13 (32.5) 11 (27.5)
8 (16.3) 23 (46.9) 12 (24.5) 6 (12.2)
40 (100.0) 0 (0.0) 0 (0.0)
48 (98.0) 1 (2.0) 0 (0.0)
1 (2.5) 17 (42.5) 11 (27.5) 4 (10.0) 3 (7.5) 2 (5.0) 2 (5.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0)
5 (10.2) 22 (44.9) 12 (24.5) 2 (4.1) 2 (4.1) 2 (4.1) 2 (4.1) 1 (2.0) 1 (2.0) 0 (0.0) 0 (0.0)
36 (90.0) 3 (7.5) 1 (2.5)
49 (100.0) 0 (0.0) 0 (0.0)
36 (90.0) 2 (5.0) 1 (2.5) 1 (2.5)
49 (100.0) 0 (0.0) 0 (0.0) 0 (0.0)
1 (2.5) 13 (32.5) 18 (45.0) 8 (20.0)
4 (8.2) 24 (49.0) 17 (35.7) 4 (8.2)
1 (2.5) 3 (7.5) 11 (27.5) 17 (42.5) 3 (7.5) 5 (12.5)
0 (0.0) 12 (24.5) 21 (42.9) 10 (20.4) 6 (12.2) 0 (0.0)
29 (72.5) 11 (27.5)
41 (83.7) 8 (16.3)
2 (5.0) 4 (10.0) 3 (7.5) 5 (12.5) 12 (30.0) 14 (35.0)
4 (8.2) 6 (12.2) 4 (8.2) 8 (16.3) 12 (24.5) 15 (30.6)
7 (18.4) 5 (13.2) 14 (36.8) 6 (15.8) 6 (15.8)
8 (18.6) 11 (25.6) 17 (39.5) 5 (11.6) 2 (4.7)
38 (95.0) 2 (5.0)
45 (93.8) 3 (6.3)
p valuea
0.03
0.36
0.72
0.08
0.16
0.13
0.01
0.04
0.96
0.35
0.80
Abbreviations: LEEP, loop electrosurgical excision procedure; OC, oral contraceptive; IUD, intrauterine device. Results presented as n (%). a Chi-squared p value for the group difference LEEP versus No LEEP.
the LEEP and No-LEEP groups respectively) between the two groups. Within-group differences from baseline to follow-up were not significant, despite a trend toward a decrease in IFN- among treated women.
3.2. IL-6 At baseline, LEEP subjects had significantly lower mean levels of IL-6 than untreated subjects
78
Table 2 Mean analysis of cytokine levels detected at baseline and six-month follow-up by LEEP status in the Cervical Composition Study, Iowa City, IA, 2007–2010. Cytokine
Follow-up
Mean within group p valuec
Not detectable, n (%)
Mean ± SDa
Range
Not detectable, n (%)
Mean ± SD
7 (17.5) 8 (16.3)
788.1 ± 513.1 882.5 ± 712.6 0.47b
164.9–2694.9 69.3–2909.8
27 (32.5) 6 (12.2)
692.1 ± 466.1 858.0 ± 817.9 0.23
0 (0.0) 0 (0.0)
111,886.0 ± 84,704.2 137,158.5 ± 111,365.0 0.24
9893.0–371,104.6 19,652.4–445,726.0
0 (0.0) 0 (0.0)
200,624.7 ± 466.1 115,595.3 ± 103,328.5 0.19
0 (0.0) 0 (0.0)
124.4 ± 118.8 150.9 ± 106.4 0.27
26.0–518.7 25.3–482.6
0 (0.0) 0 (0.0)
105.8 ± 73.6 142.2 ± 135.5 0.11
2 (5.0) 6 (12.2)
252.6 ± 193.7 331.4 ± 373.0 0.20
4.2–806.5 2.6–1304.6
5 (12.5) 7 (14.3)
203.6 ± 184.8 316.7 ± 368.0 0.06
0 (0.0) 0 (0.0)
6197.0 ± 6086.4 5359.6 ± 4824.3 0.48
305.3–30,729.5 66.4–27,495.6
0 (0.0) 0 (0.0)
13,866.0 ± 20,607.1 11,764.9 ± 32,006.7 0.70
0 (0.0) 0 (0.0)
1354.9 ± 2630.7 1004.9 ± 1578.7 0.44
22.7–12,493.8 17.4–8822.4
0 (0.0) 0 (0.0)
1104.1 ± 2056.8 1244.4 ± 2040.7 0.75
0 (0.0) 0 (0.0)
378,992.8 ± 405,930.3 334,996.5 ± 485,628.2 0.65
28,664.2–1,832,433.1 5955.8–2,908,601.4
0 (0.0) 0 (0.0)
756,088.7 ± 1,122,296.0 479,288.2 ± 574,858.2 0.16
3 (7.5) 2 (4.1)
114.5 ± 92.7 124.1 ± 109.6 0.65
4.2–411.5 12.1–435.5
5 (12.5) 4 (8.2)
0 (0.0) 0 (0.0)
5953.9 ± 5131.8 9353.9 ± 8447.0 0.02
572.0–17,445.5 565.3–40,717.3
0 (0.0) 0 (0.0)
Range 60.1–2059.6 64.2–3388.8
0.20 0.56
1115.3–2,023,863.6 11,488.9–409,327.6
0.98 0.19
8.1–368.8 14.6–664.8
0.47 0.24
3.2–710.4 2.6–1548.1
0.12 0.75
109.3–108,827.0 206.4–226,036.3
0.22 0.43
32.1–8810.5 15.9–8546.5
0.98 0.82
59,316.2–5,360,000.7 9856.7–2,875,478.7
0.11 0.17
89.4 ± 77.1 131.6 ± 166.3 0.12
2.4–381.6 7.9–852.3
0.26 0.62
5289.8 ± 5293.6 7085.7 ± 6652.3 0.16
42.8–22,611.0 257.9–29,971.8
0.29 0.11
A.F. Saftlas et al. / Journal of Reproductive Immunology 109 (2015) 74–83
EGF LEEP No-LEEP Between groups p value G-CSF LEEP No-LEEP Between groups p value GM-CSF LEEP No-LEEP Between groups p value IFN-␥ LEEP No-LEEP Between groups p value IL-1␣ LEEP No-LEEP Between groups p value IL-1 LEEP No-LEEP Between groups p value IL-1RA LEEP No-LEEP Between groups p value IL-2 LEEP No-LEEP Between groups p value IL-6 LEEP No-LEEP Between groups p value
Baseline
0 (0.0) 0 (0.0)
85,885.4 ± 83,575.0 174,818.6 ± 210,300.1 0.01
4523.9–380,668.9 3646.4–1,159,634.3
0 (0.0) 0 (0.0)
99,534.3 ± 91,033.1 133,154.9 ± 143,326.4 0.18
1 (2.5) 0 (0.0)
753.8 ± 1020.5 901.1 ± 1346.3 0.56
6.6–4656.7 28.0–7833.5
0 (0.0) 0 (0.0)
469.6 ± 577.4 1141.1 ± 3528.7 0.19
1 (2.5) 1 (2.0)
208.0 ± 207.4 214.4 ± 213.1 0.89
4.2–1173.7 8.2–979.7
3 (7.5) 2 (4.1)
154.3 ± 120.6 213.4 ± 217.5 0.11
5.0–603.5 2.6–895.6
0.24 0.85
7 (17.5) 2 (4.1)
89.2 ± 91.6 95.9 ± 83.8 0.72
3.5–387.9 12.0–403.2
7 (17.5) 4 (8.2)
63.9 ± 60.5 96.5 ± 105.4 0.07
5.0–371.5 2.6–560.9
0.43 0.57
0 (0.0) 0 (0.0)
18,230.8 ± 127,252.3 21,218.7 ± 27,398.0 0.61
751.9–118,963.1 1813.8–144,560.7
0 (0.0) 0 (0.0)
8929.1 ± 8912.3 15,642.6 ± 20,516.1 0.04
301.0–36,327.4 354.0–110,058.6
0.11 0.08
0 (0.0) 0 (0.0)
4636.8 ± 16,336.8 5035.7 ± 11,645.2 0.89
33.0–102,791.1 70.7–74,380.0
0 (0.0) 0 (0.0)
3941.8 ± 18,972.5 13,826.8 ± 50,411.9 0.21
30.2–120,433.3 48.4–290,726.7
0.30 0.94
4 (10.0) 3 (6.1)
588.8 ± 1124.7 391.1 ± 490.9 0.31
6.6–6203.0 4.0–2170.5
9 (22.5) 5 (10.2)
267.3 ± 527.2 329.5 ± 817.3 0.67
2293.4–352,857.3 7172.6–586,302.1
7.5–2956.4 11.4–24,859.8
2.4–2581.1 2.6–5695.1
0.57 0.46
0.25 0.63
0.03 0.20
A.F. Saftlas et al. / Journal of Reproductive Immunology 109 (2015) 74–83
IL-8 LEEP No-LEEP Between groups p value IL-10 LEEP No-LEEP Between groups p value IL-12 LEEP No-LEEP Between groups p value IL-13 LEEP No-LEEP Between groups p value MCP-1 LEEP No-LEEP Between groups p value RANTES LEEP No-LEEP Between groups p value TNF-␣ LEEP No-LEEP Between groups p value
LEEP, n = 40; No-LEEP, n = 49. Abbreviations: LEEP, loop electrosurgical excision procedure; SD, standard deviation; IQR, inter-quartile range. a Cytokines are presented in units of pg/ml. b t-test comparing the means between the LEEP and No-LEEP groups at baseline and at follow-up using the log-transformed cytokine values. c t-test comparing the means between baseline and follow-up cytokine values within the LEEP and No-LEEP groups using the log-transformed cytokine values.
79
80
A.F. Saftlas et al. / Journal of Reproductive Immunology 109 (2015) 74–83
Fig. 1. Box plots of the mean analyses of cytokine levels detected at baseline and six-month follow-up by LEEP status in the Cervical Composition Study, Iowa City, IA 2007–2010. Whiskers represent the minimum and maximum values detected for each cytokine. “+” represents the mean value. “*” represents statistically significant differences between the means of p < 0.10. “**” represents statistically significant differences between the means of p < 0.05. Abbreviations: BL, baseline; FU, follow-up.
(5953.9–9353.9 pg/ml, p = 0.02). Consistent with these findings, there were differences in the tertile distribution of cytokine levels at baseline between the two treatment groups (p = 0.04; Supplemental Table 1).
174,816.6 pg/ml, p = 0.01). There were also differences in the tertile distribution of IL-8 at baseline between the treatment groups, with lower levels in LEEP subjects (Supplemental Table 1).
3.3. IL-8
3.4. IL-13
As with IL-6, LEEP subjects at baseline had significantly lower levels of IL-8 than untreated women (85,885.4 versus
At baseline, the two groups had similar mean levels of IL-13. Mean levels of IL-13 at follow-up, however,
A.F. Saftlas et al. / Journal of Reproductive Immunology 109 (2015) 74–83
were lower in the LEEP group (63.9 versus 96.5 pg/ml, p = 0.07).
Table 3 Summary of correlations from the log-transformed baseline × follow-up cytokine plots.
3.5. MCP-1 As observed with IL-13 and IFN-, the mean level of MCP-1 at follow-up was significantly lower in untreated women (8929.1 versus 15,642.6 pg/ml, p = 0.04). The mean level of MCP-1 among untreated women was also lower at follow-up than at baseline (15,642.6 versus 21,218.7 pg/ml, p = 0.08). 3.6. TNF-˛ Among LEEP subjects, the mean level of TNF-˛ was significantly lower at follow-up than at baseline (588.8–267.3 pg/ml, p = 0.03); in contrast, the comparable mean levels in untreated women were unchanged (391.1–329.5 pg/ml, p = 0.20). As shown in Table 3, correlations of log-transformed cytokine levels at baseline and follow-up in untreated women were significant for all except RANTES; among LEEP subjects, significant correlations were observed for all but five cytokines: IL-1ˇ, IL-8, IL-13, MCP-1, and RANTES. Baseline and follow-up cytokine levels, however, were more highly correlated in the untreated versus treated women for seven cytokines: EGF, G-CSF, IL-2, IL-10, IL-12, IL-13, and TNF-˛. We conducted repeated measures analyses to determine if cytokine levels changed significantly over time (Supplemental Table 2). Adjusted results indicate that the LEEP and No-LEEP groups shared a common intercept for all cytokines except for IL-8 (p = 0.07) and RANTES (p = 0.03). The slopes of the LEEP and No-LEEP groups were not significantly different for any of the 16 cytokines, indicating that changes in cytokine levels over time between the two groups were not different. Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.jri.2015.01.002. 4. Discussion This study is the first to investigate the effect of LEEP on the soluble mucosal cytokine microenvironment. We evaluated changes in levels of 16 cervical cytokines in treated and untreated subjects using a well-qualified multiplex technology and assays (Chaturvedi et al., 2011; Koshiol et al., 2014) for paired longitudinal assessment. Contrary to our hypothesis, changes in cytokine levels over time of the two groups were not significantly different. In addition, LEEP did not significantly affect levels of cytokines in our samples, with the exception of TNF-˛, which decreased significantly among treated women over the study period. Overall, these findings suggest that LEEP did not substantially affect levels of the selected cytokines and chemokines with key immunological functions in the cervical cytokine microenvironment. Chronic inflammation, characterized by increased levels of TNF-˛ in cervical fluid has been associated with persistent HPV infection (Scott et al., 2013) and cervical disease
81
Pearson’s correlation
p valuea
p valueb
EGF
LEEP No-LEEP
0.47 0.74
0.002 <0.0001
0.04
G-CSF
LEEP No-LEEP
0.37 0.66
0.02 <0.0001
0.07
GM-CSF
LEEP No-LEEP
0.48 0.63
0.002 <0.0001
0.32
IFN-␥
LEEP No-LEEP
0.31 0.52
0.04 0.0001
0.25
IL-10
LEEP No-LEEP
0.36 0.71
0.02 <0.0001
0.02
IL-12
LEEP No-LEEP
0.35 0.71
0.03 <0.0001
0.02
IL-13
LEEP No-LEEP
0.18 0.69
0.28 <0.0001
0.003
IL-1␣
LEEP No-LEEP
0.43 0.51
0.005 0.0002
0.64
IL-1
LEEP No-LEEP
0.18 0.46
0.27 0.001
0.15
IL-1RA
LEEP No-LEEP
0.33 0.49
0.04 0.0004
0.38
IL-2
LEEP No-LEEP
0.32 0.63
0.04 <0.0001
0.06
IL-6
LEEP No-LEEP
0.35 0.49
0.02 0.0003
0.44
IL-8
LEEP No-LEEP
0.06 0.35
0.71 0.01
0.17
MCP-1
LEEP No-LEEP
0.12 0.36
0.45 0.01
0.25
RANTES
LEEP No-LEEP
0.17 0.16
0.31 0.28
0.96
TNF-␣
LEEP No-LEEP
0.42 0.71
0.008 <0.0001
0.04
a
p value for the Pearson’s correlation of the log-transformed cytokine values and LEEP or No-LEEP group. b p value comparing the correlations for each log-transformed cytokine value between the LEEP and No-LEEP groups using Fisher’s r-to-z transformation.
progression. We found that baseline levels of TNF-˛ tended to be higher in women who required LEEP, though this difference was not statistically significant. Following LEEP, levels of TNF-˛ decreased significantly to levels that were comparable with those among untreated subjects. Levels of TNF-˛ did not change over time in untreated subjects. One possible explanation for the decrease in the mean level of TNF-˛ in the LEEP group is that the treatment eliminated the lesions persistently infected with HPV, which may have been the source or inducer of increased baseline levels of TNF-˛ in the cervical secretions of treated women. The biological and clinical implications of these findings are unknown. Because of the small sample size, assessment of potential interactive effects between levels of TNF-˛ and other cytokines was not feasible. An unexpected observation was that mean cytokine levels in treated women, all of whom had a diagnosis of CIN2
82
A.F. Saftlas et al. / Journal of Reproductive Immunology 109 (2015) 74–83
or CIN3, tended to be lower than those among untreated women, who had CIN1 or lower. Mean cytokine levels of IFN-˛, IL-6, IL-8, IL-13, and MCP-1 were lower at baseline and/or follow-up in treated women. These results are consistent with reports of decreased levels of IFN-˛ and IL-6 in lesions of advanced cervical disease (Song et al., 2008; Rosa et al., 2012), but disagree with others where higher levels of IL-6 and IL-8 were found in cervical lavage samples from women with CIN versus those without CIN (Mhatre et al., 2012). Several environmental, genetic, and constitutional factors can cause levels of cytokines to vary widely from woman to woman, even among women of similar ages and phases of the menstrual cycle (Franklin and Kutteh, 1999). While such factors may explain the wide range of cytokine levels observed in our study, we applied strict eligibility criteria to narrow the window of cervical secretion sampling and minimize the influence of menstrual cycle timing, hormonal contraceptive use, intercourse, and genital infections. Furthermore, fluctuations in cytokine levels in paired longitudinal measurements of cervical secretions are not surprising given that the cervix is an extremely complex and dynamic environment composed of a large number of cell types that produce a wide array of cytokines, chemokines, growth factors, and anti-microbial peptides captured in cervical secretions (Rodriguez-Garcia et al., 2013). Different sampling methods have been used to measure immune markers (cervical or cervico-vaginal swabs, sponges, and lavages). Multiplexed, bead-based assays to measure immune/inflammation-related markers in cervical secretions have been widely used as a tool to measure cytokine profiles in serum and other fluids in the context of cancer and infections. We have successfully used the multiplex magnetic bead assays to identify pre-diagnostic immune/inflammatory markers of risk for a number of cancers (Purdue et al., 2013; Shiels et al., 2013); and more recently, we optimized and validated the use of Luminex bead arrays in cervical secretions (Koshiol et al., 2014). Currently, there are no standardized methods or procedures for the collection and measurement of cytokines in cervical secretions, making it difficult to compare our findings with those of other studies. We performed adjustments to our measures based on the weight of each sample to normalize for differences in volume collection between different women. Some studies using cervical lavages or sponges have not applied normalizations, making comparisons of studies a very challenging task. In addition, different assay platforms and kit manufacturers have been used in different studies, which likely have contributed to differences in results and an inability to directly compare the absolute levels of cytokines of different studies. Our study is rigorous in several other respects. All cervical secretion samples were collected among women from a single clinic, by a single practitioner, using the same clinical method and protocol, and analyzed using kits from the same batch of a single manufacturer. All study subjects recruited had colposcopy; therefore, untreated women were likely to be similar to treated women on factors that we could not adjust for analytically. We also applied a stringent set of exclusion criteria over the study period to reduce
the likelihood that other factors known to influence levels of cervical cytokines were in play. This study is also subject to some limitations. As pathology reports and HPV DNA information were unavailable, it was not possible to examine for a direct association of cytokine levels with HPV infection and cervical disease. During the study period, however, the colposcopy service at the University of Iowa uniformly treated women with a diagnosis of CIN2 or higher and did not perform a LEEP in women with CIN1 or lower. Another potential limitation could be the relatively brief follow-up period of six months. While tissue healing is complete after six months, it is possible that relevant immunological changes secondary to a LEEP may not be apparent within this relatively brief time period. Additional limitations include the relatively small sample size and the chance that some statistically significant findings are an artifact of multiple comparisons. To address some of these limitations, we are planning a larger study to investigate more directly the potential associations between cytokine alterations, persistent infection, and associated cervical disease. 5. Conclusion This exploratory study of 16 cytokines suggests that levels of TNF-˛ were significantly reduced following a LEEP, although levels of the other cytokines were not significantly changed. Because of the relatively small sample size and limited set of cytokines evaluated, further research is needed to determine if LEEP leads to dysregulation of other elements of the cervical cytokine microenvironment. Standardization of procedures and methods for collecting cervical secretions and analyzing the cervical cytokine milieu is urgently needed to allow for fair comparisons between different laboratories. Conflict of interest statement None declared. Acknowledgements This work was supported by the National Institute of Allergy and Infectious Diseases (R21-AI06811). This project has been funded in whole or in part by federal funds from the National Cancer Institute, National Institutes of Health, under Contract no. HHSN261200800001E. The content of this paper does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does the mention of trade names, commercial products, or organizations imply endorsement by the US Government. We thank Dr. Phillip Castle for contributing to the conceptualization of the project, and Mary Cherrico for coordinating and conducting the clinical aspects of this study. References Chaturvedi, A.K., Kemp, T.J., Pfeiffer, R.M., Biancotto, A., Williams, M., Munuo, S., et al., 2011. Evaluation of multiplexed cytokine and
A.F. Saftlas et al. / Journal of Reproductive Immunology 109 (2015) 74–83 inflammation marker measurements: a methodologic study. Cancer Epidemiol. Biomarkers Prev. 20, 1902–1911. Del Priore, G., Gudipudi, D.K., Montemarano, N., Restivo, A.M., Malanowska-Stega, J., Arslan, A.A., 2010. Oral diindolylmethane (DIM): pilot evaluation of a nonsurgical treatment for cervical dysplasia. Gynecol. Oncol. 116, 464–467. Franklin, R.D., Kutteh, W.H., 1999. Characterization of immunoglobulins and cytokines in human cervical mucus: influence of exogenous and endogenous hormones. J. Reprod. Immunol. 42, 93–106. Hildesheim, A., McShane, L.M., Schiffman, M., Bratti, M.C., Rodriguez, A.C., Herrero, R., et al., 1999. Cytokine and immunoglobulin concentrations in cervical secretions: reproducibility of the Weck-cel collection instrument and correlates of immune measures. J. Immunol. Methods 225, 131–143. Iwabe, T., Harada, T., Terakawa, N., 2002. Role of cytokines in endometriosis-associated infertility. Gynecol. Obstet. Invest. 53 (Suppl. 1), 19–25. Iwata, T., Fujii, T., Morii, K., Saito, M., Sugiyama, J., Nishio, H., et al., 2014. Cytokine profile in cervical mucosa of Japanese patients with cervical intraepithelial neoplasia. Int. J. Clin. Oncol., http://dx.doi.org/10. 1007/s10147-014-0680-8. Jerchel, S., Kaufhold, I., Shuchardt, L., Shima, K., Rupp, J., 2014. Host immune responses after hypoxic reactivation of IFN-gamma induced persistent chlamydia trachomatis infection. Front. Cell. Infect. Microbiol. 4, 43. Kemp, T.J., Hildesheim, A., Garcia-Pineres, A., Williams, M.C., Shearer, G.M., Rodriguez, A.C., et al., 2010. Elevated systemic levels of inflammatory cytokines in older women with persistent cervical human papillomavirus infection. Cancer Epidemiol. Biomarkers Prev. 19, 1954–1959. Koshiol, J., Sklavos, M., Wentzensen, N., Kemp, T., Schiffman, M., Dunn, S.T., et al., 2014. Evaluation of a multiplex panel of immune-related markers in cervical secretions: a methodologic study. Int. J. Cancer 134, 411–425. Kyrgiou, M., Koliopoulos, G., Martin-Hirsch, P., Arbyn, M., Prendiville, W., Paraskevaidis, E., 2006. Obstetric outcomes after conservative treatment for intraepithelial or early invasive cervical lesions: systematic review and meta-analysis. Lancet 367, 489–498. Marks, M.A., Viscidi, R.P., Chang, K., Silver, M., Burke, A., Howard, R., et al., 2011. Differences in the concentration and correlation of cervical immune markers among HPV positive and negative perimenopausal women. Cytokine 56, 798–803.
83
Mhatre, M., McAndrew, T., Carpenter, C., Burk, R.D., Einstein, M.H., Herold, B.C., 2012. Cervical intraepithelial neoplasia is associated with genital tract mucosal inflammation. Sex. Transm. Dis. 39, 591–597. Paradkar, P.H., Joshi, J.V., Mertia, P.N., Agashe, S.V., Vaidya, R.A., 2014. Role of cytokines in genesis, progression and prognosis of cervical cancer. Asian Pac. J. Cancer Prev. 15, 3851–3864. Paraskevaidis, E., Bilirakis, E., Koliopoulos, G., Lolis, E.D., Kalantaridou, S., Paschopoulos, M., et al., 2002. Cervical regeneration after diathermy excision of cervical intraepithelial neoplasia as assessed by transvaginal sonography. Eur. J. Obstet. Gynecol. Reprod. Biol. 102, 88–91. Purdue, M.P., Hofmann, J.N., Kemp, T.J., Chaturvedi, A.K., Lan, Q., Park, J.H., et al., 2013. A prospective study of 67 serum immune and inflammation markers and risk of non-Hodgkin lymphoma. Blood 122, 951–957. Rodriguez-Garcia, M., Patel, M.V., Wira, C.R., 2013. Innate and adaptive anti-HIV immune responses in the female reproductive tract. J. Reprod. Immunol. 97, 74–84. Rosa, M.I., Moraes, M.V., Vuolo, F., Petronilho, F., Bozzetti, M.C., Medeiros, L.R., et al., 2012. Association of interleukin-6 in women with persistence of DNA-HPV: a nested case-control study. Arch. Gynecol. Obstet. 285, 143–148. Sadler, L., Saftlas, A.F., Wang, W., Exeter, M., Whittaker, J., McCowan, L., 2004. Treatment for cervical intraepithelial neoplasia and risk of preterm delivery. JAMA 291, 2100–2106. Scott, M.E., Shvetsov, Y.B., Thompson, P.J., Hernandez, B.Y., Zhu, X., Wilkens, L.R., et al., 2013. Cervical cytokines and clearance of incident human papillomavirus infection: Hawaii HPV cohort study. Int. J. Cancer 133, 1187–1196. Shiels, M.S., Pfeiffer, R.M., Hildersheim, A., Engels, E.A., Kemp, T.J., Park, J.H., et al., 2013. Circulating inflammation markers and prospective risk for lung cancer. J. Natl. Cancer Inst. 105, 1871–1880. Song, S.H., Lee, J.K., Lee, N.W., Saw, H.S., Kang, J.S., Lee, K.W., 2008. Interferon-gamma (IFN-gamma): a possible prognostic marker for clearance of high-risk human papillomavirus (HPV). Gynecol. Oncol. 108, 543–548. Spracklen, C.N., Harland, K.K., Stegmann, B.J., Saftlas, A.F., 2013. Cervical surgery for cervical intraepithelial neoplasia and prolonged time to conception of a live birth: a case-control study. Br. J. Obstet. Gynaecol. 120, 960–965. Weiss, G., Goldsmith, L.T., Taylor, R.N., Bellet, D., Taylor, H.S., 2009. Inflammation in reproductive disorders. Reprod. Sci. 16, 216–229.