DeepCell Kiosk: scaling deep learning-enabled cellular image analysis with Kubernetes.

Deep learning is transforming the analysis of biological images, but applying these models to large datasets remains challenging. Here we describe the DeepCell Kiosk, cloud-native software that dynamically scales deep learning workflows to accommodate large imaging datasets. To demonstrate the scalability and affordability of this software, we identified cell nuclei in 106 1-megapixel images in ~5.5 h for ~US$250, with a cost below US$100 achievable depending on cluster configuration. The DeepCell Kiosk can be downloaded at https://github.com/vanvalenlab/kiosk-console ; a persistent deployment is available at https://deepcell.org/ .

Risk Factors for Recurrence After Surgery for Rectal Cancer in a Modern, Nationwide Population-Based Cohort.

BACKGROUND: The success of modern multimodal treatment in rectal cancer is dependent on risk prediction. Better knowledge of the risk of locoregional and distant recurrence, in relation to preoperative treatment, pathological stage, and commonly used risk factors, is needed when deciding on adjuvant therapy and surveillance. METHODS: The Swedish ColoRectal Cancer Registry was used to identify patients diagnosed with rectal adenocarcinoma between 2011 and 2018. Readily available variables, including patient, tumor, and treatment factors were exposures. Cox proportional hazard models were used to identify important risk factors for recurrence and calculate recurrence risks. RESULTS: A total of 9428 curatively resected patients were included and followed for a median of 72 months. Eighteen percent had distal recurrence and 3% had locoregional recurrence at 5 years. Risk factors with major impact on distal recurrence were pT4a (hazard ratio [HR] 5.1, 95% confidence interval [CI] 3.3-8.0), pN2b (HR 3.4, 95% CI 2.7-4.2), tumor deposit (HR 1.7, 95% CI 1.5-1.9), lymph node yield (HR 1.5, 95% CI 1.3-1.8), and tumor level 0-5 cm (HR 1.5, 95% CI 1.3-1.8). Pathologic stage and number of risk factors identified groups with markedly different recurrence risks in all neoadjuvant treatment groups. CONCLUSIONS: Readily available risk factors, as a complement to stage, are still valid and robust in all neoadjuvant treatment groups. Tumor deposit is important, while circumferential resection margin might no longer be important with improved oncological treatments and high-quality TME surgery. Tailored surveillance is possible in selected groups using risk stratification based on stage and risk factors.

Surgery for acute cholecystitis in severely comorbid patients: a population-based study on acute cholecystitis.

BACKGROUND: International guidelines recommend emergency cholecystectomy for acute cholecystitis in patients who are healthy or have mild systemic disease (ASA1-2). Surgery is also an option for patients with severe systemic disease (ASA3) in clinical practice. The study aimed to investigate the risk of complications in ASA3 patients after surgery for acute cholecystitis. METHOD: 1 634 patients treated for acute cholecystitis at three Swedish centres between 2017 and 2020 were included in the study. Data was gathered from electronic patient records and the Swedish registry for gallstone surgery, Gallriks. Logistic regression was used to assess the risk of complications adjusted for confounding factors: sex, age, BMI, Charlson comorbidity index, cholecystitis grade, smoking and time to surgery. RESULTS: 725 patients had emergency surgery for acute cholecystitis, 195 were ASA1, 375 ASA2, and 152 ASA3. Complications occurred in 9% of ASA1, 13% of ASA2, and 24% of ASA3 patients. There was no difference in 30-day mortality. ASA3 patients stayed on average 2 days longer after surgery. After adjusting for other factors, the risk of complications was 2.5 times higher in ASA3 patients than in ASA1 patients. The risk of complications after elective surgery was 5% for ASA1, 13% for ASA2 and 14% for ASA3 patients. Regardless of ASA 18% of patients treated non-operatively had a second gallstone complication within 3 months. CONCLUSION: Patients with severe systemic disease have an increased risk of complications but not death after emergency surgery. The risk is lower for elective procedures, but a substantial proportion will have new gallstone complications before elective surgery. TRIAL REGISTRATION: Not applicable.

Completeness and accuracy of the registration of recurrences in the Swedish Colorectal Cancer Registry (SCRCR) and an update of recurrence risk in colon cancer.

BACKGROUND: The completeness and accuracy of the registration of synchronous metastases and recurrences in the Swedish Colorectal Cancer Registry has not been investigated. Knowing how accurate these parameters are in the registry is a prerequisite to adequately measure the current recurrence risk. METHODS: All charts for patients diagnosed with stage I-III colorectal cancer (CRC) in two regions were reviewed. In one of the regions, all registrations of synchronous metastases were similarly investigated. After the database had been corrected, recurrence risk in colon cancer was calculated stratified by risk group as suggested by ESMO in 2020. RESULTS: In patients operated upon more than five years ago (N = 1235), there were 20 (1.6%) recurrences not reported. In more recent patients, more recurrences were unreported (4.0%). Few synchronous metastases were wrongly registered (3.6%) and, likewise, few synchronous metastases were not registered (about 1%). The five-year recurrence risk in stage II was 6% for low-risk, 11% for intermediate risk, and 23% for high-risk colon cancer patients. In stage III, it was 25% in low- and 45% in high-risk patients. Incorporation of risk factors in stage III modified the risks substantially even if this is not considered by ESMO. Adjuvant chemotherapy lowered the risk in stage III but not to any relevant extent in stage II. CONCLUSION: The registration of recurrences in the registry after 5 years is accurate to between 1 and 2% but less accurate earlier. A small number of unreported recurrences and falsely reported recurrences were discovered in the chart review. The recurrence risk in this validated and updated patient series matches what has been recently reported, except for the risk of recurrence in stage II low risk colon cancers which seem to be even a few percentage points lower (6 vs. 9%).

Beyond the NCCN Risk Factors in Colon Cancer: An Evaluation in a Swedish Population-Based Cohort.

BACKGROUND: The purpose of this study was to investigate whether pT3-4 and pN-subclassifications, lymph-node ratio (LNR), tumour deposits, pre- and postoperative carcinoembryonic antigen (CEA), and C-reactive protein (CRP)-all parameters commonly collected in clinical management-add information about recurrence risk against a background of routine clinicopathological parameters as defined by the NCCN. METHODS: The prospective cohort consisted of all 416 patients diagnosed with colon cancer stage I-III in Uppsala County between 2010 and 2015. Cox proportional hazard models were used to calculate hazard ratios for time to recurrence and overall survival. The results were compared with the entire Swedish population concerning parameters recorded in the national quality registry, SCRCR, during the same time period. RESULTS: The Uppsala cohort was representative of the entire Swedish cohort. In unadjusted analyses, pT3-subclassification, pN-subclassification, LNR, tumour deposits, elevated postoperative CEA, and preoperative CRP correlated with recurrence. After adjusting for T-, N-stage, and NCCN risk factors, pN-subclassification, sidedness, and elevated postoperative CEA levels correlated with recurrence. Survival correlated with parameters associated with recurrence, LNR, and elevated postoperative CRP. CONCLUSIONS: Additional information on recurrence risk is available from several routinely recorded parameters, but most of the risk is predicted by the commonly used clinicopathological parameters.

Recurrence Risk After Up-to-Date Colon Cancer Staging, Surgery, and Pathology: Analysis of the Entire Swedish Population.

BACKGROUND: Developments in the quality of care of patients with colon cancer have improved surgical outcome and thus the need for adjuvant chemotherapy. OBJECTIVE: To investigate the recurrence rate in a large population-based cohort after modern staging, surgery, and pathology have been implemented. DESIGN: This was a retrospective registry study. SETTINGS: Data from patients included in the Swedish Colorectal Cancer Registry covering 99% of all cases and undergoing surgery for colon cancer stages I to III between 2007 and 2012 were obtained. PATIENTS: In total, 14,325 patients who did not receive any neoadjuvant treatment, underwent radical surgery, and were alive 30 days after surgery were included. MAIN OUTCOME MEASURES: Tumor and node classification and National Comprehensive Cancer Network-defined risk factors for recurrence were used to assess overall and stage-specific 5-year recurrence rates. RESULTS: The median follow-up of nonrecurrent cases was 77 months (range, 47-118 mo). The 5-year recurrence rate was 5% in stage I, 12% in stage II, and 33% in stage III patients. In patients classified as having pT3N0 cancer with no or 1 risk factor, the 5-year recurrence rates were 9% and 11%. Risk factors for shorter time to recurrence were male sex, more advanced pT and pN classification, vascular and perineural invasion, emergency surgery, lack of central ligature, short longitudinal resection margin, postoperative complications, and, in stage III, no adjuvant chemotherapy. LIMITATIONS: The registry does not contain some recently identified factors of relevance for recurrence rates, and some late recurrences may be missing. CONCLUSIONS: The recurrence rate is less than that previously observed in historical materials, but current, commonly used risk factors are still useful in evaluating recurrence risks. Stratification by pT and pN classification and the number of risk factors enables the identification of large patient groups characterized by such a low recurrence rate that it is questionable whether adjuvant treatment is motivated. See Video Abstract at http://links.lww.com/DCR/A663.

Income disparities in loss in life expectancy after colon and rectal cancers: a Swedish register-based study.

BACKGROUND: Differences in the prognosis after colorectal cancer (CRC) by socioeconomic position (SEP) have been reported previously; however, most studies focused on survival differences at a particular time since diagnosis. We quantified the lifetime impact of CRC and its variation by SEP, using individualised income to conceptualise SEP. METHODS: Data included all adults with a first-time diagnosis of colon or rectal cancers in Sweden between 2008 and 2021. The analysis was done separately for colon and rectal cancers using flexible parametric models. For each cancer and income group, we estimated the life expectancy in the absence of cancer, the life expectancy in the presence of cancer and the loss in life expectancy (LLE). RESULTS: We found large income disparities in life expectancy after a cancer diagnosis, with larger differences among the youngest patients. Higher income resulted in more years lost following a cancer diagnosis. For example, 40-year-old females with colon cancer lost 17.64 years if in the highest-income group and 13.68 years if in the lowest-income group. Rectal cancer resulted in higher LLE compared with colon cancer. Males lost a larger proportion of their lives. All patients, including the oldest, lost more than 30% of their remaining life expectancy. Based on the number of colon and rectal cancer diagnoses in 2021, colon cancer results in almost double the number of years lost compared with rectal cancer (24 669 and 12 105 years, respectively). CONCLUSION: While our results should be interpreted in line with what individualised income represents, they highlight the need to address inequalities.

Despite multi-disciplinary team discussions the socioeconomic disparities persist in the oncological treatment of non-metastasized colorectal cancer.

BACKGROUND: The introduction of national guidelines should eliminate previously observed associations between socioeconomic status (SES) and colorectal cancer treatment. The aim of the study was to investigate whether inequalities remain. METHODS: CRCBaSe, a register-linkage originating from the Swedish Colorectal Cancer Registry, was used to identify information on patient and tumour characteristics, for 83,460 patients with stage I-III disease diagnosed 2008-2021. SES was measured as disposable income (quartiles) and the highest level of education. Outcomes of interest were emergency surgery, multidisciplinary team (MDT) conference discussion, and oncological treatment. Differences in treatment between SES groups were explored using multivariable logistic regression adjusted for year of diagnosis, age at diagnosis, sex, civil status, comorbidities, tumour location and stage. RESULTS: Patients in the highest income quartile had a lower risk of emergency surgery (OR 0.73 95%CI 0.68-0.80), a higher chance of being discussed at the preoperative (OR 1.39 95%CI 1.28-1.51) and postoperative MDT (OR 1.41 95%CI 1.30-1.53), receiving neoadjuvant (OR 1.15 95%CI 1.06-1.25) and adjuvant treatment (OR 2.04 95%CI 1.88-2.20). Higher education level increased the odds of MDT discussion but was not associated with oncological treatment. The proportion of patients discussed at the MDT increased, with almost all patients discussed since 2016. Despite this, treatment differences remained when patients diagnosed since 2016 were analysed separately. CONCLUSION: There were significant differences in how patients with different SES were treated for colorectal cancer. Further action is required to investigate the drivers of these differences as well as their impact on mortality and, ultimately, eliminate the inequalities.

Effect of the COVID-19 pandemic on the care for acute cholecystitis: a Swedish multicentre retrospective cohort study.

OBJECTIVES: The present study aimed to investigate if and how the panorama of acute cholecystitis changed in 2020 in Sweden. Seven aspects were identified, the incidence of cholecystitis, the Tokyo grade, the timing of diagnosis and treatment, the proportion treated with early surgery, the proportion of patients treated with delayed surgery, and new complications from gallstones. DESIGN: Retrospective multicentre cohort study. SETTING: 3 hospitals in Sweden, covering 675 000 inhabitants. PARTICIPANTS: 1634 patients with cholecystitis. OUTCOMES: The incidence, treatment choice and diagnostic and treatment delay were investigated by comparing prepandemic and pandemic patients. RESULTS: Patients diagnosed with cholecystitis during the pandemic were more comorbid (American Society of Anesthesiologists 2-5, 86% vs 81%, p=0.01) and more often had a diagnostic CT (67% vs 59%, p=0.01). There were variations in the number of patients corresponding with the pandemic waves, but there was no overall increase in the number of patients with cholecystitis (78 vs 76 cases/100 000 inhabitants, p=0.7) or the proportion of patients treated with surgery during the pandemic (50% vs 50%, p=0.4). There was no increase in time to admission from symptoms (both median 1 day, p=0.7), or surgery from admission (both median 1 day, p=0.9). The proportion of grades 2-3 cholecystitis was not higher during the pandemic (46% vs 44%, p=0.9). The median time to elective surgery increased (184 days vs 130 days, p=0.04), but there was no increase in new gallstone complications (35% vs 39%, p=0.3). CONCLUSION: Emergency surgery for cholecystitis was not impacted by the pandemic in Sweden. Patients were more comorbid but did not have more severe cholecystitis nor was there a delay in seeking care. Fewer patients non-operatively managed had elective surgery within 6 months of their initial diagnosis but there was no corresponding increase in gallstone complications.

Adjuvant Chemotherapy in Elderly Colorectal Cancer Patients.

The value of adjuvant chemotherapy in elderly patients has been the subject of many overviews, with opinions varying from "not effective", since randomized trials have not been performed, to "as effective as in young individuals", based upon many retrospective analyses of randomized trials that have included patients of all ages. In the absence of randomized trials performed specifically with elderly patients, retrospective analyses demonstrate that the influence on the time to tumour recurrence (TTR) may be the same as in young individuals, but that endpoints that include death for any reason, such as recurrence-free survival (RFS), disease-free survival (DFS), and overall survival (OS), are poorer in the elderly. This is particularly true if oxaliplatin has been part of the treatment. The need for adjuvant chemotherapy after colorectal cancer surgery in elderly patients is basically the same as that in younger patients. The reduction in recurrence risks may be similar, provided the chosen treatment is tolerated but survival gains are less. Adding oxaliplatin to a fluoropyrimidine is probably not beneficial in individuals above a biological age of approximately 70 years. If an oxaliplatin combination is administered to elderly patients, three months of therapy is in all probability the most realistic goal.

Recurrence Risk after Radical Colorectal Cancer Surgery-Less Than before, But How High Is It?

Adjuvant chemotherapy aims at eradicating tumour cells sometimes present after radical surgery for a colorectal cancer (CRC) and thereby diminish the recurrence rate and prolong time to recurrence (TTR). Remaining tumour cells will lead to recurrent disease that is usually fatal. Adjuvant therapy is administered based upon the estimated recurrence risk, which in turn defines the need for this treatment. This systematic overview aims at describing whether the need has decreased since trials showing that adjuvant chemotherapy provides benefits in colon cancer were performed decades ago. Thanks to other improvements than the administration of adjuvant chemotherapy, such as better staging, improved surgery, the use of radiotherapy and more careful pathology, recurrence risks have decreased. Methodological difficulties including intertrial comparisons decades apart and the present selective use of adjuvant therapy prevent an accurate estimate of the magnitude of the decreased need. Furthermore, most trials do not report recurrence rates or TTR, only disease-free and overall survival (DFS/OS). Fewer colon cancer patients, particularly in stage II but also in stage III, today display a sufficient need for adjuvant treatment considering the burden of treatment, especially when oxaliplatin is added. In rectal cancer, neo-adjuvant treatment will be increasingly used, diminishing the need for adjuvant treatment.