Serial endometrial thickness and risk of non-endometrial hormone-dependent cancers in postmenopausal women in UK Collaborative Trial of Ovarian Cancer Screening

OBJECTIVE

Estrogen is a well-established risk factor for various cancers. It causes endometrial proliferation, which is assessed routinely as endometrial thickness (ET) using transvaginal ultrasound (TVS). Only one previous study, restricted to endometrial and breast cancer, has considered ET and the risk of non-endometrial cancer. The aim of this study was to explore the association between baseline and serial ET measurements and nine non-endometrial hormone-sensitive cancers, in postmenopausal women, using contemporary statistical methodology that attempts to minimize the biases typical of endogenous serial data.

METHODS

This was a cohort study nested within the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS). In the ultrasound arm of UKCTOCS, 50639 postmenopausal women, aged 50-74, underwent annual TVS examination, of whom 38 105 had a valid ET measurement, no prior hysterectomy and complete covariate data, and were included in this study. All women were followed up through linkage to national cancer registries. The effect of ET on the risk of six estrogen-dependent cancers (breast, ovarian, colorectal, bladder, lung and pancreatic) was assessed using joint models for longitudinal biomarker and time-to-event data, and Cox models were used to assess the association between baseline ET measurement and these six cancers in addition to liver cancer, gastric cancer and non-Hodgkin's lymphoma (NHL). All models were adjusted for current hormone-replacement therapy (HRT) use, body mass index, age at last menstrual period, parity and oral contraceptive pill use.

RESULTS

The 38 105 included women had a combined total of 267 567 (median, 8; interquartile range, 5-9) valid ET measurements. During a combined total of 407 838 (median, 10.9) years of follow-up, 1398 breast, 351 endometrial, 381 lung, 495 colorectal, 222 ovarian, 94 pancreatic, 79 bladder, 62 gastric, 38 liver cancers and 52 NHLs were registered. Using joint models, a doubling of ET increased significantly the risk of breast (hazard ratio (HR), 1.21; 95% CI, 1.09-1.36; P = 0.001), ovarian (HR, 1.39; 95% CI, 1.06-1.82; P = 0.018) and lung (HR, 1.25; 95% CI, 1.02-1.54; P = 0.036) cancers. There were no statistically significant associations between ET and the remaining six cancers.

CONCLUSION

Postmenopausal women with high/increasing ET on TVS are at increased risk of breast, ovarian and lung cancer. It is important that clinicians are aware of these risks, as TVS is a common investigation. © 2019 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of the International Society of Ultrasound in Obstetrics and Gynecology.

Sonographers' self-reported visualization of normal postmenopausal ovaries on transvaginal ultrasound is not reliable: results of expert review of archived images from UKCTOCS

OBJECTIVE

In the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS), self-reported visualization rate (VR) of the ovaries by the sonographer on annual transvaginal sonographic (TVS) examinations was a key quality control (QC) metric. The objective of this study was to assess self-reported VR using expert review of a random sample of archived images of TVS examinations from UKCTOCS, and then to develop software for measuring VR automatically.

METHODS

A single expert reviewed images archived from 1000 TVS examinations selected randomly from 68 931 TVS scans performed in UKCTOCS between 2008 and 2011 with ovaries reported as 'seen and normal'. Software was developed to identify the exact images used by the sonographer to measure the ovaries. This was achieved by measuring caliper dimensions in the image and matching them to those recorded by the sonographer. A logistic regression classifier to determine visualization was trained and validated using ovarian dimensions and visualization data reported by the expert.

RESULTS

The expert reviewer confirmed visualization of both ovaries (VR-Both) in 50.2% (502/1000) of the examinations. The software identified the measurement image in 534 exams, which were split 2:1:1 providing training, validation and test data. Classifier mean accuracy on validation data was 70.9% (95% CI, 70.0-71.8%). Analysis of test data (133 exams) provided a sensitivity of 90.5% (95% CI, 80.9-95.8%) and specificity of 47.5% (95% CI, 34.5-60.8%) in detecting expert confirmed visualization of both ovaries.

CONCLUSIONS

Our results suggest that, in a significant proportion of TVS annual screens, the sonographers may have mistaken other structures for normal ovaries. It is uncertain whether or not this affected the sensitivity and stage at detection of ovarian cancer in the ultrasound arm of UKCTOCS, but we conclude that QC metrics based on self-reported visualization of normal ovaries are unreliable. The classifier shows some potential for addressing this problem, though further research is needed. © 2017 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of the International Society of Ultrasound in Obstetrics and Gynecology.

Time to diagnosis of Type I or II invasive epithelial ovarian cancers: a multicentre observational study using patient questionnaire and primary care records

OBJECTIVE

To compare time to diagnosis of the typically slow-growing Type I (low-grade serous, low-grade endometrioid, mucinous, clear cell) and the more aggressive Type II (high-grade serous, high-grade endometrioid, undifferentiated, carcinosarcoma) invasive epithelial ovarian cancer (iEOC).

DESIGN

Multicentre observational study.

SETTING

Ten UK gynaecological oncology centres.

POPULATION

Women diagnosed with primary EOC between 2006 and 2008.

METHODS

Symptom data were collected before diagnosis using patient questionnaire and primary-care records. We estimated patient interval (first symptom to presentation) using questionnaire data and diagnostic interval (presentation to diagnosis) using primary-care records. We considered the impact of first symptom, referral and stage on intervals for Type I and Type II iEOC.

MAIN OUTCOME MEASURES

Patient and diagnostic intervals.

RESULTS

In all, 78% of 60 Type I and 21% of 134 Type II iEOC were early-stage. Intervals were comparable and independent of stage [e.g. median patient interval for Type I: early-stage 0.3 months (interquartile range 0.3-3.0) versus late-stage 0.3 months (interquartile range 0.3-4.5), P = 0.8]. Twenty-seven percent of women with Type I and Type II had diagnostic intervals of at least 9 months. First symptom (questionnaire) was also similar, except for the infrequent abnormal bleeding (Type I 15% versus Type II 4%, P = 0.01). More women with Type I disease (57% versus 41%, P = 0.04) had been referred for suspected gynaecological cancer. Median time from referral to diagnosis was 1.4 months for women with iEOC referred via a 2-week cancer referral to any specialty compared with 2.6 months (interquartile range 2.0-3.7) for women who were referred routinely to gynaecology.

CONCLUSION

Overall, shorter diagnostic delays were seen when a cancer was suspected, even if the primary tumour site was not recognised to be ovarian. Despite differences in carcinogenesis and stage for Type I and Type II iEOC, time to diagnosis and symptoms were similar. Referral patterns were different, implying subtle symptom differences. If symptom-based interventions are to impact on ovarian cancer survival, it is likely to be through reduced volume rather than stage-shift. Further research on histological subtypes is needed.

TWEETABLE ABSTRACT

No difference in time to diagnosis for Type I versus Type II invasive epithelial ovarian cancers.

Quality assurance and its impact on ovarian visualization rates in the multicenter United Kingdom Collaborative Trial of Ovarian Cancer Screening (UKCTOCS)

OBJECTIVE

To describe the quality assurance (QA) processes and their impact on visualization of postmenopausal ovaries in the ultrasound arm of a multicenter screening trial for ovarian cancer.

METHODS

In the United Kingdom Collaborative Trial of Ovarian Cancer Screening, 50 639 women aged 50-74 years were randomized to the ultrasound arm and underwent annual transvaginal ultrasound (TVS) examinations. QA processes were developed during the course of the trial and included regular monitoring of the visualization rate (VR) of the right ovary. Non-subjective factors identified previously as impacting on VR of the right ovary were included in a generalized estimating equation model for binary outcomes to enable comparison of observed vs adjusted VR between individual sonographers who had undertaken > 1000 scans during the trial and comparison between centers. Observed and adjusted VRs of sonographers and centers were ranked according to the highest VR. Analysis of annual VRs of sonographers and those of the included centers was undertaken.

RESULTS

Between June 2001 and December 2010, 48 230 of 50 639 women attended one of 13 centers for a total of 270 035 annual TVS scans. One or both ovaries were seen in 228 145 (84.5%) TVS scans. The right ovary was seen on 196 426 (72.7%) of the scans. For the 78 sonographers included in the model, the median difference between observed and adjusted VR was -0.7% (range, -7.9 to 5.9%) and the median change in VR rank after adjustment was 3 (range, 0-18). For the 13 centers, the median difference between observed and adjusted VR was -0.5% (range, -2.2 to 1%), with no change in ranking after adjustment. The median adjusted VR was 73% (interquartile range (IQR), 65-82%) for sonographers and 74.7% (IQR, 67.1-79.0%) for centers. Despite the increasing age of the women being scanned, there was a steady decrease in the number of sonographers with VR < 60% (21.4% in 2002 vs 2.0% in 2010) and an increase in sonographers with VR > 80% (14.3% in 2002 vs 40.8% in 2010). The median VR of the centers increased from 65.5% (range, 55.7-81.0%) in 2001 to 80.3% (range, 74.5-90.9%) in 2010.

CONCLUSIONS

A robust QA program can improve visualization of postmenopausal ovaries and is an essential component of ultrasound-based ovarian cancer screening trials. While VR should be adjusted for non-subjective factors that impact on ovarian visualization, subjective factors are likely to be the largest contributors to differences in VR.