Artificial intelligence for detection of microsatellite instability in colorectal cancer-a multicentric analysis of a pre-screening tool for clinical application.

BACKGROUND: Microsatellite instability (MSI)/mismatch repair deficiency (dMMR) is a key genetic feature which should be tested in every patient with colorectal cancer (CRC) according to medical guidelines. Artificial intelligence (AI) methods can detect MSI/dMMR directly in routine pathology slides, but the test performance has not been systematically investigated with predefined test thresholds. METHOD: We trained and validated AI-based MSI/dMMR detectors and evaluated predefined performance metrics using nine patient cohorts of 8343 patients across different countries and ethnicities. RESULTS: Classifiers achieved clinical-grade performance, yielding an area under the receiver operating curve (AUROC) of up to 0.96 without using any manual annotations. Subsequently, we show that the AI system can be applied as a rule-out test: by using cohort-specific thresholds, on average 52.73% of tumors in each surgical cohort [total number of MSI/dMMR = 1020, microsatellite stable (MSS)/ proficient mismatch repair (pMMR) = 7323 patients] could be identified as MSS/pMMR with a fixed sensitivity at 95%. In an additional cohort of N = 1530 (MSI/dMMR = 211, MSS/pMMR = 1319) endoscopy biopsy samples, the system achieved an AUROC of 0.89, and the cohort-specific threshold ruled out 44.12% of tumors with a fixed sensitivity at 95%. As a more robust alternative to cohort-specific thresholds, we showed that with a fixed threshold of 0.25 for all the cohorts, we can rule-out 25.51% in surgical specimens and 6.10% in biopsies. INTERPRETATION: When applied in a clinical setting, this means that the AI system can rule out MSI/dMMR in a quarter (with global thresholds) or half of all CRC patients (with local fine-tuning), thereby reducing cost and turnaround time for molecular profiling.

Investigating the poor outcomes of BRAF-mutant advanced colorectal cancer: analysis from 2530 patients in randomised clinical trials.

Background: To improve strategies for the treatment of BRAF-mutant advanced colorectal cancer (aCRC) patients, we examined individual data from patients treated with chemotherapy alone in three randomised trials to identify points on the treatment pathway where outcomes differ from BRAF wild-types. Patients and methods: 2530 aCRC patients were assessed from three randomised trials. End-points were progression-free survival, response rate, disease control rate, post-progression survival (P-PS) and overall survival. Treatments included first-line oxaliplatin/fluorouracil (OxFU) and second-line irinotecan. Clinicians were unaware of BRAF-status. Results: 231 patients (9.1%) had BRAF-mutant tumours. BRAF-mutation conferred significantly worse survival independent of associated clinicopathological factors known to be prognostic. Compared with wild-type, BRAF-mutant patients treated with first-line OxFU had similar DCR (59.2% versus 72%; adjusted OR = 0.76, P = 0.24) and PFS (5.7 versus 6.3 months; adjusted HR = 1.14, P = 0.26). Following progression on first-line chemotherapy, BRAF-mutant patients had a markedly shorter P-PS (4.2 versus 9.2 months, adjusted HR = 1.69, P < 0.001). Fewer BRAF-mutant patients received second-line treatment (33% versus 51%, P < 0.001), but BRAF-mutation was not associated with inferior second-line outcomes (RR adjusted OR = 0.56, P = 0.45; PFS adjusted HR = 1.01, P = 0.93). Significant clinical heterogeneity within the BRAF-mutant population was observed: a proportion (24.3%) had good first-line PFS and P-PS (both >6 months; OS = 24.0 months); however, 36.5% progressed rapidly through first-line chemotherapy and thereafter, with OS = 4.7 months. Conclusions: BRAF-mutant aCRC confers a markedly worse prognosis independent of associated clinicopathological features. Chemotherapy provides meaningful improvements in outcome throughout treatment lines. Post-progression survival is markedly worse and vigilance is required to ensure appropriate delivery of treatment after first-line progression.

A feasibility study testing four hypotheses with phase II outcomes in advanced colorectal cancer (MRC FOCUS3): a model for randomised controlled trials in the era of personalised medicine?

BACKGROUND: Molecular characteristics of cancer vary between individuals. In future, most trials will require assessment of biomarkers to allocate patients into enriched populations in which targeted therapies are more likely to be effective. The MRC FOCUS3 trial is a feasibility study to assess key elements in the planning of such studies. PATIENTS AND METHODS: Patients with advanced colorectal cancer were registered from 24 centres between February 2010 and April 2011. With their consent, patients' tumour samples were analysed for KRAS/BRAF oncogene mutation status and topoisomerase 1 (topo-1) immunohistochemistry. Patients were then classified into one of four molecular strata; within each strata patients were randomised to one of two hypothesis-driven experimental therapies or a common control arm (FOLFIRI chemotherapy). A 4-stage suite of patient information sheets (PISs) was developed to avoid patient overload. RESULTS: A total of 332 patients were registered, 244 randomised. Among randomised patients, biomarker results were provided within 10 working days (w.d.) in 71%, 15 w.d. in 91% and 20 w.d. in 99%. DNA mutation analysis was 100% concordant between two laboratories. Over 90% of participants reported excellent understanding of all aspects of the trial. In this randomised phase II setting, omission of irinotecan in the low topo-1 group was associated with increased response rate and addition of cetuximab in the KRAS, BRAF wild-type cohort was associated with longer progression-free survival. CONCLUSIONS: Patient samples can be collected and analysed within workable time frames and with reproducible mutation results. Complex multi-arm designs are acceptable to patients with good PIS. Randomisation within each cohort provides outcome data that can inform clinical practice.

KRAS mutation analysis on low percentage of colon cancer cells: the importance of quality assurance.

KRAS mutation testing is mandatory for patients with metastatic colorectal cancer who are eligible for treatment with an epidermal growth factor receptor targeting agent, since tumors with a mutation are not sensitive to the drug. Several methods for mutation testing are in use and the need for external quality assurance has been demonstrated. An often little addressed but important issue in external quality assurance schemes is a low percentage of tumor cells in the test samples, where the analytical sensitivity of most tests becomes critical. Using artificial samples based on a mixture of cell lines with known mutation status of the KRAS gene, we assessed the reliability of a series of commonly used methods (Sanger sequencing, high resolution melting, pyrosequencing, and amplification refractory mutation system-polymerase chain reaction) on samples with 0, 2.5, 5, 10, and 15 % mutated cells. Nine laboratories throughout Europe participated and submitted a total of ten data sets. The limit of detection of each method differed, ranging from >15-5 % tumor cells. All methods showed a decreasing correct mutation call rate proportionally with decreasing percentage of tumor cells. Our findings indicate that laboratories and clinicians need to be aware of the decrease in correct mutation call rate proportionally with decreasing percentage of tumor cells and that external quality assurance schemes need to address the issue of low tumor cell percentage in the test samples.

KRAS and BRAF mutations in Nigerian colorectal cancers.

PURPOSE: Activation of the KRAS oncogene is implicated in colorectal carcinogenesis and mutations have been reported in 30-50% of cases. BRAF mutation, though less common, is also reported and importantly associated with shorter progression-free interval. This study aims to determine the KRAS and BRAF mutation statuses of Nigerian colorectal cancers (CRC). METHODS: Mutation analysis was carried out on archival paraffin-embedded blocks of CRC tissues. KRAS codons 12, 13 and 61 and BRAF V600E were assessed by pyrosequencing after DNA extraction from 200 cases at the Leeds Institute of Molecular Medicine, St. James's University Hospital, UK. Mutation rates and the spectra were determined. RESULTS: Pyrosequencing was successful in 112 of 200 cases. KRAS mutation in codons 12 and 13 was demonstrated in 23 of 112 cases (21%); none in codon 61. BRAF mutation in codon 600 was demonstrated in 4.5%. CONCLUSION: This study shows that 21% of Nigerian CRC patients carry a KRAS mutation; half the rate in Caucasians; and that BRAF mutation also occurs in Nigerian CRC cancers.

What can the molecular pathologist offer for optimal decision making?

As a consequence of new innovative therapies and therapeutic combinations, the treatment of advanced colorectal cancer is becoming increasingly complex. Validated molecular biomarkers could contribute to patient management, but until recently, none has been routinely used. With the recognition that activating mutations of the KRAS oncogene can predict resistance to anti-epidermal growth factor receptor agents, the clinical value of biomarkers in advanced colorectal cancer has been brought to the fore. Prognostic and predictive biomarkers that reflect the molecular and therapeutic complexities of advanced colorectal cancer may provide valuable information regarding likely clinical outcome and therapeutic response and thus may improve patient management and therapeutic agent selection. Such biomarkers are discussed herein.

Superselective transcatheter embolization of traumatic renal hemorrhage.

Selective angiography is an established diagnostic modality in acute renal trauma [1-3]. Renal hemorrhage, once demonstrated by angiography, can then be treated by arterial embolization [4-5]. This paper describes the successful transcatheter management of a renal laceration in a patient with multiple injuries secondary to an abdominal gunshot wound. Adominal exploration had established that the renal hemorrhage could not be surgically controlled without performing a nephrectomy.

Abnormal pulmonary gallium accumulation in P. carinii pneumonia.

The potentially fatal interstitial pneumonia of Pneumocystis carinii is a frequent opportunistic invader of patients treated for malignancy with immunosuppressive and cytotoxic agents. Generalized Ga-67 pulmonary localization which is markedly disproportionate to the clinical and radiographic findings has led to earlier diagnoses by open lung biopsy in this setting. The potential usefulness of gallium scintigraphy in patients with suspected P. Carini pneumonia is presented.

Mediastinoscopy: a diagnostic aid in metastatic carcinoma of the breast.

Six patients with metastatic carcinoma of the breast underwent mediastinoscopy to obtain tissue for estrogen receptor analysis and pathologic confirmation of metastatic tumor. The indication for mediastioxcopy was an abnormal mediastinal accumulation of gallium in five patients, only two of whom had an abnormality noted on tomography. All six patients had tumor recovered by mediastinoscopy as demonstrated by pathologic examination and/or estrogen receptor assay, Therefore, the diagnosis of mediastinal metastases in breast cancer may be suggested by either chest roentgenograms, mediastinal tomography, or gallium scintigraphy. Mediastinoscopy is a safe, effective procedure capable of establishing the diagnosis of metastatic carcinoma of the breast and of procuring sufficient tissue for estrogen receptor analysis in patients without more readily accessible sites of metastases.

67Ga accumulation in pulmonary lesions associated with bleomycin toxicity.

The radionuclide, gallium-67, accumulates in a wide variety of neoplastic and inflammatory lesions. Diffuse bilateral lung localization of 67Ga occurred in two cases of interstitial pneumonitis associated with bleomycin therapy. Clinical symptoms and laboratory evaluation of pulmonary status correlated with scintigraphic studies in these two patients, while discrepancies between scintigraphic and roentgenographic findings were observed.

67Ga radionuclide imaging in Burkitt's lymphoma.

Whole-body gallium imaging was evaluated as a diagnostic tool for 14 patients with Burkitt's lymphoma, and a dramatic gallium affinity for the tumor was demonstrated. The gallium image is recommended as a staging procedure and as an effective test for determining the therapeutic response.

Usefulness of gallium scintigraphy in primary and metastatic breast carcinoma.

Gallium-67-citrate whole-body scintigraphy was incorporated into the evaluation of 125 patients with primary and metastatic breast carcinoma. A prospective analysis of these patients demonstrated gallium scanning to be limited in its corroborative value and of practical use only in detecting metastatic breast carcinoma in the mediastinum. Since the extent of tumor involvement is a significant prognostic factor, the gallium scan merits consideration for the detection of mediastinal metastases.

Breast scintigraphy with 99mTc-pertechnetate and 67Ga-citrate.

Radionuclide breast scintigraphy was evaluated as a noninvasive tumor-localizing modality. Technetium-99m-pertechnetate (99mTcO4) demonstrated good correlation between malignancy and postive scintigraphy (88% accuracy in 16 cases of breast carcinoma). The high false-positive rate (29% of proven benign breast disease) limits the use of 99mTcO4 as an aid to differential diagnosis. Gallium-67-citrate (67Ga) is limited as a diagnostic adjunct (localizing in only five of ten breast malignancies). Refined techniques of positioning, shielding, gamma camera imaging, and computer assistance have helped in visualizing abnormal radionuclide accumulation.

Renal localization of gallium-67 citrate.

Renal localization of 67Ga citrate has been observed in 1.7% of 2,000 whole-body gallium scans in three major diseases: lymphoreticular neoplasms, leukemia, and malignant melanoma. Correlation of this finding with autopsy data in 13 cases revealed 9 with tumor and 2 with an inflammatory process. Findings in two cases were normal. In the remaining patients, correlation with abnormal urograms and laboratory data was poor. Nevertheless, gallium localization in the kidneys may be the first indicator of renal involvement with either tumor or inflammation.

Radionuclide studies in Hodgkin's disease and lymphomas.

A rational, multidisciplinary approach to Hodgkin's disease and the non-Hodgkin's lymphomas has been responsible for major advances in therapy. Invasive diagnostic procedures and exploratory laparotomy, with their associated complications, make nontraumatic radionuclide imaging most appealing in both the clinical staging of disease and in evaluating therapy. Gallium-67-citrate, the tumor scanning agent of the early 1970's, has demonstrated a marked affinity for Hodgkin's disease and the other lymphomas. False positives are few, with sensitivity greater than 70% throughout the spectrum of Hodgkin's disease and the histiocytic lymphomas. In addition to confirming sites of suspected neoplasm, this agent has proved useful in the detection of occult involvement. Moreover, resolution of abnormal gallium-67 concentrations on follow-up studies functions as a visual ancillary index of therapeutic response. The value of wholebody gallium-67 scintigraphy is further enhanced when used in conjunction with routine technetium brain, bone, liver, and spleen scans. While the diagnostic accuracy of gallium-67 studies has been limited in the abdomen due to bowel activity, our attempts to improve these results with the tumor-seeking radiopharmaceutical indium-111-Bleomycin were unrewarding and subsequently were discontinued. Finally, radionuclide lymphography has also been explored. Its diagnostic usefulness in detecting pelvic and abdominal lymph node involvement warrants further investigation.