Deep learning to predict breast cancer sentinel lymph node status on INSEMA histological images.

BACKGROUND: Sentinel lymph node (SLN) status is a clinically important prognostic biomarker in breast cancer and is used to guide therapy, especially for hormone receptor-positive, HER2-negative cases. However, invasive lymph node staging is increasingly omitted before therapy, and studies such as the randomised Intergroup Sentinel Mamma (INSEMA) trial address the potential for further de-escalation of axillary surgery. Therefore, it would be helpful to accurately predict the pretherapeutic sentinel status using medical images. METHODS: Using a ResNet 50 architecture pretrained on ImageNet and a previously successful strategy, we trained deep learning (DL)-based image analysis algorithms to predict sentinel status on hematoxylin/eosin-stained images of predominantly luminal, primary breast tumours from the INSEMA trial and three additional, independent cohorts (The Cancer Genome Atlas (TCGA) and cohorts from the University hospitals of Mannheim and Regensburg), and compared their performance with that of a logistic regression using clinical data only. Performance on an INSEMA hold-out set was investigated in a blinded manner. RESULTS: None of the generated image analysis algorithms yielded significantly better than random areas under the receiver operating characteristic curves on the test sets, including the hold-out test set from INSEMA. In contrast, the logistic regression fitted on the Mannheim cohort retained a better than random performance on INSEMA and Regensburg. Including the image analysis model output in the logistic regression did not improve performance further on INSEMA. CONCLUSIONS: Employing DL-based image analysis on histological slides, we could not predict SLN status for unseen cases in the INSEMA trial and other predominantly luminal cohorts.

Preexisting musculoskeletal burden and its development under letrozole treatment in early breast cancer patients.

One of the most common adverse events (AEs) occurring during treatment with aromatase inhibitors (AIs) is musculoskeletal pain. The aim of our study was to analyze the influence of preexisting muscle/limb pain and joint pain on the development of AI-induced musculoskeletal AEs. Women eligible for upfront adjuvant endocrine therapy with letrozole were included in the PreFace study, a multicenter phase IV trial. During the first treatment year, they were asked to record musculoskeletal AEs monthly by answering questions regarding pain symptoms and rating the pain intensity on a numeric rating scale from 0 (no pain) to 10 (very strong pain). Pain values were compared using nonparametric statistical tests. Overall, 1,416 patients were evaluable. The average pain value over all time points in women with preexisting muscle/limb pain was 4.3 (median 4.3); in those without preexisting pain, it was 2.0 (median 1.7). In patients without preexisting muscle/limb pain, pain levels increased relatively strongly within the first 6 months (mean increase +0.9, p < 0.00001) in comparison with those with preexisting pain (mean increase +0.3, p < 0.001), resulting in a statistically significant difference (p < 0.00001) between the two groups. The development of joint pain was similar in the two groups. Women without preexisting muscle/limb pain or joint pain have the greatest increase in pain after the start of adjuvant AI therapy. Women with preexisting pain have significantly higher pain values. The main increase in pain values takes place during the first 6 months of treatment.

Association between breast cancer risk factors and molecular type in postmenopausal patients with hormone receptor-positive early breast cancer.

PURPOSE: Evidence shows that genetic and non-genetic risk factors for breast cancer (BC) differ relative to the molecular subtype. This analysis aimed to investigate associations between epidemiological risk factors and immunohistochemical subtypes in a cohort of postmenopausal, hormone receptor-positive BC patients. METHODS: The prospective, single-arm, multicenter phase IV PreFace study (Evaluation of Predictive Factors Regarding the Effectivity of Aromatase Inhibitor Therapy) included 3529 postmenopausal patients with hormone receptor-positive early BC. Data on their epidemiological risk factors were obtained from patients' diaries and their medical histories. Data on estrogen receptor, progesterone receptor, and HER2 receptor status were obtained from pathology reports. Patients with incomplete information were excluded. Data were analyzed using conditional inference regression analysis, analysis of variance, and the chi-squared test. RESULTS: In a cohort of 3392 patients, the strongest association with the molecular subtypes of BC was found for hormone replacement therapy (HRT) before diagnosis of early BC. The analysis showed that patients who took HRT at diagnosis had luminal A-like BC more often (83.7%) than those who had never taken HRT or had stopped taking it (75.5%). Luminal B-like BC and HER2-positive BC were diagnosed more often in women who had never taken HRT or had stopped taking it (13.3% and 11.2%, respectively) than in women who were taking HRT at diagnosis of BC (8.3% and 8.0%, respectively). CONCLUSIONS: This analysis shows an association between HRT and the distribution of molecular subtypes of BC. However, no associations between other factors (e.g., age at diagnosis, body mass index, smoking status, age at menopause, number of deliveries, age at first delivery, breastfeeding history, or family history) were noted.

Fumarate hydratase (FH) deficiency in uterine leiomyomas: recognition by histological features versus blind immunoscreening.

Hereditary leiomyomatosis and renal cell carcinoma (HLRCC) syndrome is a rare autosomal dominant disease caused by germline mutations in the fumarate hydratase (FH) gene. Affected individuals develop cutaneous and uterine leiomyomas and aggressive RCC. To date, only few publications described the frequency and morphology of FH-deficient uterine leiomyomas. We reviewed 22 cases collected over 8 years from routine and consultation files based on distinctive histological features. In addition, we screened 580 consecutive uterine leiomyomas from 484 patients, 23 extra-uterine and 8 uterine leiomyosarcomas, and 6 leiomyomas with bizarre nuclei for FH loss using immunohistochemistry (IHC) on tissue microarrays (TMAs). All 22 FH-deficient cases were suspected on H&E sections and confirmed by FH IHC. Patients' ages ranged from 25 to 70 years (median 36). Seventeen patients had multiple nodules (2-14) measuring up to 11.8 cm. None of the patients had stigmata or family history of the HLRCC syndrome. Histologically, all FH-deficient tumors showed consistent and reproducible features as reported previously. FH loss was detected in 2/534 evaluable leiomyomas (0.4%), but in none of leiomyosarcomas. Two of six leiomyomas with bizarre nuclei were FH-deficient. FH-deficient uterine leiomyomas are rare in routine material (= 0.4%). They can be reliably identified or suspected by consistent morphological features. Our data showed predictive morphology to be superior to blind IHC screening for detecting them. The relationship of FH-deficient uterine smooth muscle tumors to the HLRCC syndrome needs further clarification.

Sagittal magnetic resonance imaging-scan orientation significantly influences accuracy of femoral posterior condylar offset measurement.

BACKGROUND: In total knee arthroplasty the femoral posterior condylar offset (PCO) may serve as a potential branch for correct femoral component positioning. The technique of adjusting the sagittal magnetic resonance imaging (MRI)-scan on which it is measured has not been investigated in previous literature, but may be subject to variances due to knee joint positioning or axial localizer scan angulation. The purpose of this study was to investigate the effect of simulated femur rotation on the accuracy of PCO measurement. MATERIALS AND METHODS: Ten asymptomatic knee joints underwent MRI investigations. A sagittal plane perpendicular to the transepicondylar axis was defined as the true-sagittal plane (tsP). Sagittal images were reformatted in the tsP and angulated by 5° and - 5° in medial and lateral direction. In total each knee received three scans in 0°, 5° and - 5° axial localizer scan angulation. Medial and lateral PCO measurement was performed in each MRI-scan angulation. RESULTS: Simulated external rotation decreased medial PCO size by 1.7 mm (95% CI 0.5994-3.127) (p = 0.012), and simulated internal rotation increased medial PCO size by 2.1 mm (95% CI 1.142-2.994) (p = 0.001). Lateral PCO size increased by 1.9 mm (95% CI 0.5660-3.412) and decreased by 2.1 mm (95% CI 1.142-2.994) with simulated external and internal rotation, respectively (p = 0.011; p = 0.0007). CONCLUSION: This study shows the high sensitivity of medial and lateral PCO measurements to small changes of MRI axial localizer scan angulations simulating minor degrees of internal or external femur rotation. Thus, absolute PCO values should be interpreted with caution if the sagittal image acquisition is not standardized.

Implementation and Feasibility of Electronic Patient-Reported Outcome (ePRO) Data Entry in the PRAEGNANT Real-Time Advanced and Metastatic Breast Cancer Registry.

PURPOSE: Patient-reported outcomes (PROs) have been incorporated into clinical trials for many symptoms and medical conditions. A transition from paper-based capture of PROs to electronic PROs (ePROs) has recently started. This study reports on the feasibility of ePRO assessment in a prospective registry including molecular data for patients with advanced breast cancer. METHODS: As part of the PRAEGNANT network, patients were invited by clinical trial staff, physicians, and nurses to complete three standardized Internet-based questionnaires (EQ 5D 5 L, CES-D and IPAQ). Feasibility was assessed by the staff members who assigned the user accounts by the patients. The completeness of the questionnaires was also assessed. RESULTS: Fifteen of 17 patients who were asked agreed to participate to complete the PRO questionnaires (EQ-5D-5L and CES-D). However, the IPAQ (physical activity) questionnaire was only validly completed by 9 patients. Feasibility was ranked better by the physicians and dedicated clinical trial staff than by the nursing staff. CONCLUSIONS: Incorporating ePRO questionnaires into an advanced breast cancer registry is feasible, and no major hurdles were reported. Involving stakeholders from the start, the application is tailored to the capacities and abilities of both patients and clinical staff. The patients' compliance was better with some questionnaires, but others may present difficulties.

Characteristics of femorotibial joint geometry in the trochlear dysplastic femur.

The medial and lateral tibia plateau geometry has been linked with the severity of trochlear dysplasia. The aim of the present study was to evaluate the tibial slope and the femoral posterior condylar offset in a cohort of consecutive subjects with a trochlear dysplastic femur to investigate whether the condylar offset correlates with, and thus potentially compensates for, tibial slope asymmetry. Magnetic resonance imaging was used to assess the severity of trochlear dysplasia as well as the tibial slope and posterior offset of the femoral condyles separately for the medial and lateral compartment of the knee joint in 98 subjects with a trochlear dysplastic femur and 88 control subjects. A significant positive correlation was found for the medial tibial slope and the medial posterior condylar offset in the study group (r(2) = 0.1566; P < 0.001). This relationship was significant for all subtypes of trochlear dysplasia and was most pronounced in the severe trochlear dysplastic femur (Dejour type D) (r(2) = 0.3734; P = 0.04). No correlation was found for the lateral condylar offset and the lateral tibial slope in the study group or for the condylar offset and the tibial slope on both sides in the control group. The positive correlation between the medial femoral condylar offset and the medial tibial slope, that is, a greater degree of the medial tibial slope indicated a larger offset of the medial femoral condyle, appears to represent a general anthropomorphic characteristic of distal femur geometry in patients with a trochlear dysplastic femur.