A federated approach to identify women with early-stage cervical cancer at low risk of lymph node metastases

OBJECTIVE

Lymph node metastases (pN+) in presumed early-stage cervical cancer negatively impact prognosis. Using federated learning, we aimed to develop a tool to identify a group of women at low risk of pN+, to guide the shared decision-making process concerning the extent of lymph node dissection.

METHODS

Women with cervical cancer between 2005 and 2020 were identified retrospectively from population-based registries: the Danish Gynaecological Cancer Database, Swedish Quality Registry for Gynaecologic Cancer and Netherlands Cancer Registry. Inclusion criteria were: squamous cell carcinoma, adenocarcinoma or adenosquamous carcinoma; The International Federation of Gynecology and Obstetrics 2009 IA2, IB1 and IIA1; treatment with radical hysterectomy and pelvic lymph node assessment. We applied privacy-preserving federated logistic regression to identify risk factors of pN+. Significant factors were used to stratify the risk of pN+.

RESULTS

We included 3606 women (pN+ 11%). The most important risk factors of pN+ were lymphovascular space invasion (LVSI) (odds ratio [OR] 5.16, 95% confidence interval [CI], 4.59-5.79), tumour size 21-40 mm (OR 2.14, 95% CI, 1.89-2.43) and depth of invasion>10 mm (OR 1.81, 95% CI, 1.59-2.08). A group of 1469 women (41%)-with tumours without LVSI, tumour size ≤20 mm, and depth of invasion ≤10 mm-had a very low risk of pN+ (2.4%, 95% CI, 1.7-3.3%).

CONCLUSION

Early-stage cervical cancer without LVSI, a tumour size ≤20 mm and depth of invasion ≤10 mm, confers a low risk of pN+. Based on an international privacy-preserving analysis, we developed a useful tool to guide the shared decision-making process regarding lymph node dissection.

Accuracy of colposcopy in the Swedish screening program

INTRODUCTION

Sensitivity and specificity of colposcopy vary greatly between studies and efficacy in clinical studies seldom corresponds with effectiveness in a real-life setting. It is unclear whether colposcopists' experience affects assessment; studies show divergent results. The study's objective was to investigate the accuracy of colposcopies in the Swedish screening program, the variability in colposcopists' assessments and whether degree of experience affects accuracy in a routine setting.

MATERIAL AND METHODS

Cross-sectional register study. All colposcopic assessments with a concomitant histopathological sample from women aged at least 18 years, performed between 1999 and September 2020 in Sweden. The main outcome measure was accuracy. The accuracy of colposcopic assessments was calculated as overall agreement with linked biopsies, with three outcomes: Normal vs Atypical, Normal vs Low-Grade Atypical vs High-Grade Atypical, and Non-High-Grade Atypical vs High-Grade Atypical. A time-trend analysis was performed. The accuracy of identifiable colposcopists related to experience was analyzed.

RESULTS

In total, 82 289 colposcopic assessments with linked biopsies were included for analysis of the outcome Normal vs Atypical; average accuracy was 63%. Overrating colposcopic findings was four times more common than underrating. No time trend in accuracy was noted during the study period. Accuracy in distinguishing High-Grade from Non-High-Grade lesions was better: 76%. Among identifiable colposcopists, overall accuracy was 67%. Some had significantly better accuracy than others, but no correlation with experience was found.

CONCLUSIONS

Colposcopy, including in a referral setting, has low accuracy in distinguishing Normal from Atypical. Increased experience alone does not lead to improvement. This is supported by the substantial differences in performance between colposcopists.

Colposcopic assessment by Swedescore, evaluation of effectiveness in the Swedish screening programme: a cross-sectional study

OBJECTIVE

To evaluate the effectiveness and performance of Swedescore in the Swedish screening programme.

DESIGN

Cross-sectional register study.

SETTING AND POPULATION

All Swedish women aged over 18 years with a colposcopic assessment linked to a biopsy in the Swedish National Cervical Screening Registry, 2015-20.

METHODS

Colposcopies with Swedescore were compared with the histopathological diagnosis of cervical intraepithelial neoplasia grade 2 or higher (CIN2+). The respective influence of cytology and human papillomavirus (HPV) testing, at referral for colposcopy and concurrently with colposcopy, were investigated in regression models.

MAIN OUTCOME MEASURES

CIN2+.

RESULTS

A total of 11 317 colposcopic assessments with Swedescore were included. Odds ratios for CIN2+ increased for every step in the Swedescore scale. At Swedescore ≥0-1, the proportion of CIN2+ was 9.8%. At Swedescore ≥8, the specificity was 93.3% and the positive predictive value was 60.1%, Area under the receiver operating characteristics curve (AUC) was 0.71. If the smear had been abnormal at referral, a normal colposcopy (Swedescore 0-1) was still associated with a CIN2+ risk of more than 5%. In the regression model, cytology and HPV had higher odds ratio for CIN2+ than colposcopy; the combination resulted in an AUC of 0.88.

CONCLUSIONS

Swedescore works well in a routine clinical setting but colposcopy assessed with Swedescore was inferior to that reported in previous clinical studies. No safe cutoff level was identified for refraining from biopsy. See-and-treat at Swedescore 8-10 is feasible only if referral cytology showed high-grade squamous intraepithelial lesion.

TWEETABLE ABSTRACT

No safe cutoff level for refraining from biopsy nor for see-and-treat with Swedescore.

Increased Institutional Surgical Experience in Robot-Assisted Radical Hysterectomy for Early Stage Cervical Cancer Reduces Recurrence Rate: Results from a Nationwide Study

The aim of this study was to evaluate the impact of institutional surgical experience on recurrence following robotic radical hysterectomy (RRH) for early stage cervical cancer. All women in Sweden who underwent an RRH for stage IA2-IB1 cervical cancer at tertiary referral centers from its implementation in December 2005 until June 2017 were identified using a Swedish nationwide register and local hospital registers. Registry data were controlled by a chart review of all women. Recurrence rates and patterns of recurrence were compared between early and late (≤50 vs. >50 procedures) institutional series. Six hundred and thirty-five women were included. Regression analysis identified a lower risk of recurrence with increased experience but without a clear cut off level. Among the 489 women who did not receive adjuvant radio chemotherapy (RC-T), the rate of recurrence was 3.6% in the experienced cohort (>50 procedures) compared to 9.3% in the introductory cohort (p < 0.05). This was also seen in tumors < 2 cm regardless of RC-T (p < 0.05), whereas no difference in recurrence was seen when analyzing all women receiving RC-T. In conclusion, the rate of recurrence following RRH for early stage cervical cancer decreased with increased institutional surgical experience, in tumors < 2 cm and in women who did not receive adjuvant RC-T.

Risk of vaginal cancer among hysterectomised women with cervical intraepithelial neoplasia: a population-based national cohort study

OBJECTIVE

To study the risk of vaginal cancer among hysterectomised women with and without cervical intraepithelial neoplasia (CIN).

DESIGN

Population-based national cohort study.

SETTING AND POPULATION

All Swedish women, 5 million in total, aged 20 and up, 1987-2011 using national registries.

METHODS

The study cohort was subdivided into four exposure groups: hysterectomised with no previous history of CIN3 and without prevalent CIN at hysterectomy; hysterectomised with a history of CIN3/adenocarcinoma in situ (AIS); hysterectomised with prevalent CIN at hysterectomy; non-hysterectomised.

MAIN OUTCOME MEASURE

Vaginal cancer.

RESULTS

We identified 898 incident cases of vaginal cancer. Women with prevalent CIN at hysterectomy and those with a history of CIN3/AIS had incidence rates (IR) of vaginal cancer of 51.3 (95% CI 34.4-76.5) and 17.1 (95% CI 12.5-23.4) per 100 000, respectively. Age-adjusted IR-ratios (IRRs) compared with hysterectomised women with benign cervical history were 21.0 (95% CI 13.4-32.9) and 5.81 (95% CI 4.00-8.43), respectively. IR for non-hysterectomised women was 0.87 (95% CI 0.81-0.93) and IRR was 0.37 (95% CI 0.30-0.46). In hysterectomised women with prevalent CIN, the IR remained high after 15 years of follow up: 65.7 (95% CI 21.2-203.6).

CONCLUSIONS

Our findings suggest that hysterectomised women with prevalent CIN at surgery should be offered surveillance. Hysterectomised women without the studied risk factors have a more than doubled risk of contracting vaginal cancer compared with non-hysterectomised women in the general population. Still, the incidence rate does not justify screening.

TWEETABLE ABSTRACT

High risk of contracting vaginal cancer among hysterectomised women having prevalent CIN at surgery.

No survival difference between robotic and open radical hysterectomy for women with early-stage cervical cancer: results from a nationwide population-based cohort study

PURPOSE

The aim of the study was to compare overall survival (OS) and disease-free survival (DFS) after open and robotic radical hysterectomy for early-stage cervical cancer.

PATIENTS AND METHODS

This was a nationwide population-based cohort study on all women with cervical cancer stage IA1-IB of squamous, adenocarcinoma or adenosquamous histological subtypes, from January 2011 to December 2017, for whom radical hysterectomy was performed. The Swedish Quality Register of Gynaecologic Cancer was used for identification. To ensure quality and conformity of data and to disclose patients not yet registered, hospital registries were reviewed and validated. Cox and propensity score regression analysis and univariable and multivariable regression analysis were performed in regard to OS and DFS.

RESULTS

There were 864 women (236 open and 628 robotic) included in the study. The 5-year OS was 92% and 94% and DFS was 84% and 88% for the open and robotic cohorts, respectively. The recurrence pattern was similar in both groups. Using propensity score analysis and matched cohorts of 232 women in each surgical group, no significant differences were seen in survival: 5-year OS of 92% in both groups (hazard ratio [HR], 1.00; 95% confidence interval [CI], 0.50-2.01) and DFS of 85% vs 84% in the open and robotic cohort, respectively (HR, 1.08; 95% CI, 0.66-1.78). In univariable and multivariable analysis with OS as the end-point, no significant factors were found, and in regard to DFS, tumour size (p < 0.001) and grade 3 (p = 0.02) were found as independent significant risk factors.

CONCLUSION

In a complete nationwide population-based cohort, where radical hysterectomy for early-stage cervical cancer is highly centralised, neither long-term survival nor pattern of recurrence differed significantly between open and robotic surgery.

Effect of Fee on Cervical Cancer Screening Attendance--ScreenFee, a Swedish Population-Based Randomised Trial

BACKGROUND

Attendance in the cervical cancer screening programme is one of the most important factors to lower the risk of contracting the disease. Attendance rates are often low in areas with low socioeconomic status. Charging a fee for screening might possibly decrease attendance in this population. Screening programme coverage is low in low socio-economic status areas in Gothenburg, Sweden, but has increased slightly after multiple interventions in recent years. For many years, women in the region have paid a fee for screening. We studied the effect of abolishing this fee in a trial emanating from the regular cervical cancer screening programme.

METHOD

Individually randomised controlled trial. All 3 124 women in three low-resource areas in Gothenburg, due for screening during the study period, were randomised to receive an offer of a free test or the standard invitation stating the regular fee of 100 SEK (≈11 €). The study was conducted during the first six months of 2013. Attendance was defined as a registered Pap smear within 90 days from the date the invitation was sent out.

RESULTS

Attendance did not differ significantly between women who were charged and those offered free screening (RR 0.93; CI 0.85-1.02). No differences were found within the districts or as an effect of age, attendance after the most recent previous invitation or previous experience of smear taking.

CONCLUSION

Abolishment of a modest screening fee in socially disadvantaged urban districts with low coverage, after previous multiple systematic interventions, does not increase attendance in the short term. Other interventions might be more important for increasing attendance in low socio-economic status areas.

TRIAL REGISTRATION

ClinicalTrials.gov NCT02378324.

A 6-month regimen for EPTB with intermittent treatment in the continuation phase: a study of 679 cases

BACKGROUND

No large studies have demonstrated the effectiveness of a 6-month regimen for all forms of extra-pulmonary tuberculosis (EPTB).

METHODS

Retrospective, observational analysis to evaluate the effectiveness of a 6-month treatment regimen (2HRZE/4H3R3) for all patients diagnosed with EPTB in Caracas, Venezuela, from 1 January 1998 to 31 December 2000, using direct observation.

RESULTS

Of 679 patients enrolled, 101 (14.9%) had AIDS. In 83.2% the diagnosis was based on microbiological, histological or genetic amplification information. Of 612 (90.1%) patients who took more than 90% of the doses, treatment had to be altered in six (1%) due to drug side effects. Of the remaining 606 patients who took more than 90% of the doses, 603 (99.5%) were cured and three failed. In the follow-up conducted 2 years after the end of treatment, only 6 relapsed (1%). Cures without relapse were achieved in 24 cases of central nervous system involvement, 27 cases of osteoarticular involvement and in the 42 who had miliary and/or disseminated TB.

CONCLUSION

A 6-month treatment regimen for all forms of EPTB, with treatment three times a week during the continuation phase, was highly effective.

Epidemiological features of rotavirus infection in Caracas, Venezuela: implications for rotavirus immunization programs

The epidemiological features of rotavirus infection may be quite relevant for evaluation of the performance of a rotavirus vaccine in different settings, as well as for monitoring its impact during vaccination under routine conditions. This article describes some important issues regarding rotavirus epidemiology in Venezuela, where major field trials of rotavirus vaccine have been carried out. Rotaviruses was significantly more frequently observed in inpatient (43%) than in outpatient (21%) consultations for diarrhea in infants and young children. There was a high prevalence of rotavirus illness, regardless of socioeconomic conditions, but the risk of dehydration was greater among the lower socioeconomic groups. Rotavirus disease occurs year-round, with a slight seasonal pattern. Eighty-five percent of rotavirus-positive diarrheal episodes, as well as 86% of cases of dehydration due to rotavirus, occurred during the first year of life. However, rotavirus illnesses occur less commonly during the first months of life (0-2 months), which may be a result of protection by transplacental antibodies. The pattern of acquisition of rotavirus antibody was consistent with this age distribution of disease and with optimal age for vaccination. Thus, regional epidemiological characteristics of rotavirus infection may affect optimal performance of rotavirus vaccine.

Evaluation of the antigenicity and reactogenicity of varying formulations of the rhesus rotavirus-based quadrivalent and the M37 rotavirus vaccine candidates

Three phase I trials of the rhesus rotavirus (RRV)-based quadrivalent vaccine [composed of serotype 3 (RRV), and serotypes 1 (D x RRV), 2 (DS1 x RRV), and 4 (ST3 x RRV) human rotavirus x RRV reassortants] and the M37 (nursery strain) rotavirus vaccine candidates were conducted in an attempt to find a safe and optimally antigenic formulation. Infants 10-20 weeks old received, in trial I, 1) the quadrivalent vaccine as two separate bivalent doses (1 x 10(4) PFU each of D x RRV and RRV, followed 4 weeks later by 1 x 10(4) PFU each of DS1 x RRV and ST3 x RRV) or 2) placebo; in trial II, 1) one dose of quadrivalent vaccine (10(4) PFU of each component), or 2) two doses of quadrivalent vaccine, or 3) a 10(4) PFU dose of M37 vaccine, or 4) M37 vaccine followed by the quadrivalent vaccine, or 5) placebo; in trial III, 1) a dose of a higher-titered quadrivalent vaccine (10(5) PFU of each component), or 2) two doses of higher titered quadrivalent vaccine, or 3) a dose of higher titered M37 vaccine (10(5) PFU) or 4) two doses of M37 vaccine (10(5) PFU), or 5) M37 vaccine (10(5) PFU) followed by the higher titered quadrivalent vaccine, or 6) placebo. A mild, transient fever during the first week postvaccination was associated with the bivalent or quadrivalent vaccines but not with the M37 vaccine. Fourfold or greater serum IgA ELISA responses to rotavirus were observed in 48-92% of the infants receiving quadrivalent vaccine and in 32-50% of those receiving M37 vaccine.(ABSTRACT TRUNCATED AT 250 WORDS)

Reactogenicity and immunogenicity of a high-titer rhesus rotavirus-based quadrivalent rotavirus vaccine

We evaluated the reactogenicity and antigenicity of a quadrivalent rotavirus vaccine composed of serotype 3 rhesus rotavirus (RRV) and three single-gene-substitution reassortants of RRV and human strain D (D x RRV, serotype 1), DS1 (DS1 x RRV, serotype 2), or ST3 (ST3 x RRV, serotype 4) in a double-masked study with 302 infants in Caracas, Venezuela. Three doses of the quadrivalent vaccine composed of either 10(5) PFU (low titer) or 10(6) PFU (high titer) of each component were administered to 99 and 101 infants, respectively, at 4-week intervals starting at the second month of age; 102 infants received a placebo. Postvaccination reactions were monitored by home visits every other day during the week postvaccination. The vaccine was associated with the occurrence of mild, short-lived febrile episodes in 26 and 23% of the recipients after the first doses of high- or low-titer vaccine, respectively, in comparison with 13% of the infants receiving the placebo. Febrile reactions occurred less frequently in vaccinees after the second or third dose than after the initial dose. The vaccine was not significantly associated with diarrhea or any additional symptom or sign. Serum specimens obtained shortly before the first, 4 weeks after the first, and 4 weeks after the third dose of vaccine or placebo were tested by an immunoglobulin A enzyme-linked immunosorbent assay and by neutralization assays. Seroresponses occurred significantly more often after 3 doses than after a single dose of either vaccine. Immunoglobulin A responses were observed in 80 and 79% of the infants after 3 doses of high- or low-titer vaccine, respectively. Most of the infants tested developed a neutralization response to RRV after 3 doses of the high- (90%) or low-(88%) titer vaccine. Neutralization response rates to human rotavirus serotypes 1 to 4 after 3 doses were similar in both vaccine and 87 of 90 receiving the high-titer vaccine developed seroresponses, as detected by any of the assays employed. The study indicates that 3 doses of quadrivalent vaccine at a titer of 10(6) PFU of each component offered no advantage over the lower-titer preparation for use in efficacy trials.