Breast cancer incidence, stage distribution, and treatment shifts during the 2020 COVID-19 pandemic: a nationwide population-level study

BACKGROUND

The first COVID-19 wave in 2020 necessitated temporary suspension of non-essential medical services including organized cancer screening programs in Belgium. This study assessed the impact of the pandemic on breast cancer (BC) incidence, stage at diagnosis, and management in Belgium in 2020.

METHODS

All Belgian residents diagnosed with in situ or invasive BC in 2015-2020 in the nationwide, population-based cancer registry database were included. Incidence trends for 2015-2019 were extrapolated to predict incidence and stage distribution for 2020 and compared with the observed values. National healthcare reimbursement data were used to examine treatment strategies. Exact tumor diameter and nodal involvement, extracted from pathology reports, were analyzed for 2019 and 2020.

RESULTS

74,975 tumors were selected for analysis of incidence and clinical stage. Invasive BC incidence declined by -5.0% in 2020, with a drop during the first COVID-19 wave (Mar-Jun; -23%) followed by a rebound (Jul-Dec; +7%). Predicted and observed incidence (in situ + invasive) was not different in patients < 50 years. In the 50-69 and 70 + age groups, significant declines of -4.1% and - 8.4% respectively were found. Excess declines were seen in clinical stage 0 and I in Mar-Jun, without excess increases in clinical stage II-IV tumors in Jul-Dec. There was no increase in average tumor diameter or nodal involvement in 2020. Patients diagnosed in Mar-Jun received significantly more neoadjuvant therapy, particularly neoadjuvant hormonal therapy for patients with clinical stage I-II BC.

CONCLUSIONS

BC incidence decline in 2020 in Belgium was largely restricted to very early-stage BC and patients aged 50 and over. Delayed diagnosis did not result in an overall progression to higher stage at diagnosis in 2020. Observed treatment adaptations in Belgium were successful in prioritizing patients for surgery while preventing tumor progression in those with surgical delay. Continuation of monitoring BC incidence and stage in the future is crucial.

Hospital Volume for Rectal Cancer Resection Plays a Pivotal Role in Improving Outcomes: A Population-based Analysis

OBJECTIVE

Assessing the association between hospital surgical volume (SV) and outcomes after rectal cancer surgery (RCS), using national population-based data.

SUMMARY BACKGROUND DATA

For RCS, the association of higher SV with improved short- and/or long-term outcomes remains controversial.

METHODS

National cancer registry data and administrative data were used to select patients diagnosed with stage I-III rectal cancer in 2009-2018 and who underwent RCS. The average annual SV of RCS was categorised as low (<15; LV), medium (15-29; MV) or high (≥30; HV). The association between SV and 90-day and 1-year excess postoperative mortality (90DPM and 1YEPM) and 5-year observed survival (5YOS) was evaluated.

RESULTS

From the 11,519 patients , RCS was performed in LV (4,088; 36%), MV (2,795; 24%) or HV (4,636; 40%) hospitals. Observed 90DPM was significantly better in HV (2.3% 95%CI[1.9,2.8]) than in LV (3.7% 95%CI[3.2,4.4]) and MV (3.5% 95%CI[2.9,4.3]) with adjusted OR 1.4, P<0.0001. Continuous regression analysis showed significantly higher 90DPM in annual SV <35 compared to ≥35 (OR 1.6 95%CI[1.21,2.11]; P=0.0009). Observed 1YEPM was significantly better in HV (2.9% 95%CI[2.2,3.6]) compared to LV (4.7% 95%CI [3.9,5.6]) with adjusted excess HR 1.31 95%CI[1.00,1.73] and P=0.05, and to MV (5.0% 95%CI[4.0,6.1]) with adjusted excess HR 1.45 95%CI[1.09,1.94] and P=0.01. The 5YOS was significantly better in HV (75.9% 95%CI[74.6,77.2]) than in LV (70.3% 95%CI[68.8,71.8]) and MV (71.5% 95%CI[69.7,73.2]) with adjusted HR 1.4 in both LV and MV versus HV, P≤0.003.

CONCLUSIONS

This population-based study identified robustly superior outcomes at 90-days, 1-year and 5-years after RCS in hospitals with higher volumes.

Surgery of the primary tumor in patients with de novo metastatic breast cancer: a nationwide population-based retrospective cohort study in Belgium

PURPOSE

We aimed to assess the impact of surgery of primary tumor in overall survival (OS) of women with de novo metastatic breast cancer.

METHODS

Nationwide, population-based retrospective cohort study of women diagnosed with de novo metastatic breast cancer in Belgium, between Jan/2010-Dec/2014. Data was obtained from the Belgian Cancer Registry and administrative databases. "Surgery" group was defined by surgery of primary tumor up to nine months after diagnosis. We excluded women who did not receive systemic treatment or did not complete nine months follow-up after diagnosis. All the subsequent analyses reporting on overall survival and the stratified outcome analyses were performed based on this nine-month landmark cohort. OS was estimated using Kaplan-Meier method and compared using adjusted Cox proportional hazards models controlling for confounders with 95% confidence intervals (CI). We performed a stratified analysis according to surgery timing and a propensity score matching analysis.

RESULTS

1985 patients, 534 (26.9%) in the "Surgery" and 1451 (73.1%) in the "No Surgery" group. Patients undergoing surgery were younger (p < 0.001), had better performance status (PS) (p < 0.001), and higher proportion of HER2-positive and triple-negative breast cancer (p = 0.012). Median follow-up was 86.0 months (82.6-88.5). Median OS was 60.1 months (57.1-68.2) in the "Surgery" vs. 41.9 months (39.8-44.2) in the "No Surgery" group (adjusted HR 0.56; 0.49-0.64). OS was similar when surgery was performed upfront or after systemic treatment. Propensity score matching analysis confirmed the same findings.

CONCLUSION

Among patients receiving systemic treatment for de novo metastatic breast cancer and surviving nine months or more, those who received surgery of the primary tumor within nine months of diagnosis have longer subsequent survival than those who did not.

Trends and variation in treatment of early breast cancer in European certified breast centres: an EUSOMA-based analysis

BACKGROUND

Practice indicators (PI) measure provided care making use of real-world data. This study describes trends and variations in adjuvant treatment of early breast cancer (EBC) using the European Society of Breast Cancer Specialists (EUSOMA) database.

METHODS

The analysis was conducted on anonymous cumulative data registered by 56 certified breast centres, which all entered at least 500 new diagnoses in the database in the 10-year period 2010-2019. Practice trends of radiotherapy, endocrine treatment, chemotherapy, and anti-HER2 therapy were evaluated. The association with age group (<50, 50-69, ≥70) and geographical area of the centre (Northern, Central, Southern Europe; NE, CE, SE) was assessed with the Pearson Chi2 test for independence in contingency tables.

RESULTS

In total, 150,150 patients with EBC were selected. Overall, radiotherapy was administered more frequently in NE centres, and conversely, endocrine, chemo-, and anti-HER2 therapy were used more frequently in SE centres (p<0.001). 46.9% of the pN1 patients received postmastectomy radiotherapy, with significant differences by age and geographical region (p < 0.001). Adjuvant endocrine treatment for endocrine-sensitive carcinoma in situ was administered in 46.1%, with a decreasing trend during the study period (58.5-34.5%; p < 0.001). Anti-HER2 therapy was delivered in 75.6% of all patients with HER2BC T1a/bN0, patients older than 70 received anti-HER2 in 67.6% in SE compared to 31.3% in NE centres.

CONCLUSION

Important variations in EBC management between European certified breast centres have been demonstrated. PI using real-world data can help to monitor, evaluate, and eventually guide and align good clinical practice in the management of breast cancer.

A Population-Based Study Using Belgian Cancer Registry Data Supports Centralization of Esophageal Cancer Surgery in Belgium

BACKGROUND

Esophageal cancer surgery outcomes benefit from higher hospital volumes. Despite the evidence, organization of national health care often is complex and depends on various factors. The volume-outcome results of this population-based study supported national health policy measures regarding concentration of esophageal resections in Belgium.

METHODS

The Belgian Cancer Registry (BCR) database was linked to administrative data on cancer treatment. All Belgian patients with newly diagnosed esophageal cancer in 2008-2018 undergoing resection were allocated to the hospital at which surgery was performed. The study assessed hospital volume association with 90-day mortality and 5-year overall survival, classifying average annual hospital volume of resections as low (LV, <6), medium (MV, 6-19), or high (HV, ≥20) and as a continuous covariate in the regression models.

RESULTS

The study included 4156 patients who had surgery in 79 hospitals (2 HV hospitals [37% of all surgeries], 12 MV hospitals [30% of all surgeries], and 65 LV hospitals [33% of all surgeries]). Adjusted 90-day mortality in HV hospitals was lower than in LV hospitals (odds ratio [OR], 0.37; 95% CI, 0.21-0.65; p = 0.001). Case-mix adjusted 5-year survival was superior in HV versus LV (hazard ratio [HR], 0.43; 95% CI, 0.31-0.60; p < 0.001). The continuous model demonstrated a lower 90-day mortality (OR, 0.40; 95% CI, 0.23-0.71; p = 0.002) and a superior 5-year survival (HR, 0.45; 95% CI, 0.33-0.63; p < 0.001) in hospitals with volumes of 40 or more resections annually.

CONCLUSION

Population-based data from the BCR confirmed a strong volume-outcome association for esophageal resections. Improved 5-year survival in centers with annual volumes of 20 or more resections was driven mainly by the achievement of superior 90-day mortality. These findings supported centralization of esophageal resections in Belgium.

Machine Learning-Based Extraction of Breast Cancer Receptor Status From Bilingual Free-Text Pathology Reports

As part of its core business of gathering population-based information on new cancer diagnoses, the Belgian Cancer Registry receives free-text pathology reports, describing results of (pre-)malignant specimens. These reports are provided by 82 laboratories and written in 2 national languages, Dutch or French. For breast cancer, the reports characterize the status of estrogen receptor, progesterone receptor, and Erb-b2 receptor tyrosine kinase 2. These biomarkers are related with tumor growth and prognosis and are essential to define therapeutic management. The availability of population-scale information about their status in breast cancer patients can therefore be considered crucial to enrich real-world scientific studies and to guide public health policies regarding personalized medicine. The main objective of this study is to expand the data available at the Belgian Cancer Registry by automatically extracting the status of these biomarkers from the pathology reports. Various types of numeric features are computed from over 1,300 manually annotated reports linked to breast tumors diagnosed in 2014. A range of popular machine learning classifiers, such as support vector machines, random forests and logistic regressions, are trained on this data and compared using their F₁ scores on a separate validation set. On a held-out test set, the best performing classifiers achieve F₁ scores ranging from 0.89 to 0.92 for the four classification tasks. The extraction is thus reliable and allows to significantly increase the availability of this valuable information on breast cancer receptor status at a population level.

Proposal for magnetic resonance imaging-guided salvage radiotherapy for prostate cancer

BACKGROUND

A subset of patients experience a biochemical recurrence following radical prostatectomy. Radiotherapy can salvage those patients, provided that all disease is encompassed within the target volume. We hypothesized that this can be achieved more adequately with magnetic resonance imaging (MRI)-guided treatment planning.

MATERIAL AND METHODS

From January 2009 to April 2014, 183 patients were referred to our department for salvage radiotherapy (SRT). According to protocol, patients received a planning computed tomography (CT) as well as an MRI in treatment position. All MRI scans were retrospectively reviewed by an experienced uro-radiologist.

RESULTS

Median prostate-specific antigen (PSA) value at time of referral was 0.3 ng/ml (range 0.02-4.7 ng/ml). MRI did not show any suspected macroscopic disease in 137 patients (75%). In 46 (25%) patients, MRI did indicate a pelvic recurrence. The mean PSA level was significantly higher in patients with a suspected recurrence on MRI (0.4 vs. 1.4 ng/ml, p < .001) on a Student's t-test. The mean follow-up was 33 months (range 5-69 months). Biochemical disease-free survival (bDFS) was significantly worse in patients with suspected disease on MRI [hazard ratio (HR) 2.9, p < .0001]. bDFS was significantly worse in the subgroup where the macroscopic recurrences on MRI received a lower radiation dose (HR 3.4, p = .01).

CONCLUSION

MRI detects loco-regional disease in a substantial subset of patients with a biochemical recurrence after prostatectomy, especially in a PSA above 0.5 μg/l. Lack of MRI-based dose escalation on these macroscopic recurrences could explain some of the biochemical progression observed after SRT.