Interpretable artificial intelligence to optimise use of imatinib after resection in patients with localised gastrointestinal stromal tumours: an observational cohort study.

BACKGROUND: Current guidelines recommend use of adjuvant imatinib therapy for many patients with gastrointestinal stromal tumours (GISTs); however, its optimal treatment duration is unknown and some patient groups do not benefit from the therapy. We aimed to apply state-of-the-art, interpretable artificial intelligence (ie, predictions or prescription logic that can be easily understood) methods on real-world data to establish which groups of patients with GISTs should receive adjuvant imatinib, its optimal treatment duration, and the benefits conferred by this therapy. METHODS: In this observational cohort study, we considered for inclusion all patients who underwent resection of primary, non-metastatic GISTs at the Memorial Sloan Kettering Cancer Center (MSKCC; New York, NY, USA) between Oct 1, 1982, and Dec 31, 2017, and who were classified as intermediate or high risk according to the Armed Forces Institute of Pathology Miettinen criteria and had complete follow-up data with no missing entries. A counterfactual random forest model, which used predictors of recurrence (mitotic count, tumour size, and tumour site) and imatinib duration to infer the probability of recurrence at 7 years for a given patient under each duration of imatinib treatment, was trained in the MSKCC cohort. Optimal policy trees (OPTs), a state-of-the-art interpretable AI-based method, were used to read the counterfactual random forest model by training a decision tree with the counterfactual predictions. The OPT recommendations were externally validated in two cohorts of patients from Poland (the Polish Clinical GIST Registry), who underwent GIST resection between Dec 1, 1981, and Dec 31, 2011, and from Spain (the Spanish Group for Research in Sarcomas), who underwent resection between Oct 1, 1987, and Jan 30, 2011. FINDINGS: Among 1007 patients who underwent GIST surgery in MSKCC, 117 were included in the internal cohort; for the external cohorts, the Polish cohort comprised 363 patients and the Spanish cohort comprised 239 patients. The OPT did not recommend imatinib for patients with GISTs of gastric origin measuring less than 15·9 cm with a mitotic count of less than 11·5 mitoses per 5 mm2 or for those with small GISTs (<5·4 cm) of any site with a count of less than 11·5 mitoses per 5 mm2. In this cohort, the OPT cutoffs had a sensitivity of 92·7% (95% CI 82·4-98·0) and a specificity of 33·9% (22·3-47·0). The application of these cutoffs in the two external cohorts would have spared 38 (29%) of 131 patients in the Spanish cohort and 44 (35%) of 126 patients in the Polish cohort from unnecessary treatment with imatinib. Meanwhile, the risk of undertreating patients in these cohorts was minimal (sensitivity 95·4% [95% CI 89·5-98·5] in the Spanish cohort and 92·4% [88·3-95·4] in the Polish cohort). The OPT tested 33 different durations of imatinib treatment (<5 years) and found that 5 years of treatment conferred the most benefit. INTERPRETATION: If the identified patient subgroups were applied in clinical practice, as many as a third of the current cohort of candidates who do not benefit from adjuvant imatinib would be encouraged to not receive imatinib, subsequently avoiding unnecessary toxicity on patients and financial strain on health-care systems. Our finding that 5 years is the optimal duration of imatinib treatment could be the best source of evidence to inform clinical practice until 2028, when a randomised controlled trial with the same aims is expected to report its findings. FUNDING: National Cancer Institute.

Blue Ribbon Committee I Review: Findings and Impact.

OBJECTIVE: Review the subsequent impact of recommendations made by the 2004 American Surgical Association Blue Ribbon Committee (BRC I) Report on Surgical Education. BACKGROUND: Current leaders of the American College of Surgeons and the American Surgical Association convened an expert panel to review the impact of the BRC I report and make recommendations for future improvements in surgical education. METHODS: BRC I members reviewed the 2004 recommendations in light of the current status of surgical education. RESULTS: Some of the recommendations of BRC I have gained traction and have been implemented. There is a well-organized national curriculum and numerous educational offerings. There has been greater emphasis on preparing faculty to teach and there are ample opportunities for professional advancement as an educator. The number of residents has grown, although not at a pace to meet the country's needs either by total number or geographic distribution. The number of women in the profession has increased. There is greater awareness and attention to resident (and faculty) well-being. The anticipated radical change in the educational scheme has not been adopted. Training in surgical research still depends on the resources and interests of individual programs. Financing student and graduate medical education remains a challenge. CONCLUSIONS: The medical landscape has changed considerably since BRC I published its findings in 2005. A contemporary assessment of surgical education and training is needed to meet the future needs of the profession and our patients.

Patient metabolic profile defined by liver and muscle 18F-FDG PET avidity is independently associated with overall survival in gastric cancer.

BACKGROUND: PET-CT-based patient metabolic profiling is a novel concept to incorporate patient-specific metabolism into gastric cancer care. METHODS: Staging PET-CTs, demographics, and clinicopathologic variables of gastric cancer patients were obtained from a prospectively maintained institutional database. PET-CT avidity was measured in tumor, liver, spleen, four paired muscles, and two paired fat areas in each patient. The liver to rectus femoris (LRF) ratio was defined as the ratio of SUVmean of liver to the average SUVmean of the bilateral rectus femoris muscles. Kaplan-Meier and Cox-proportional hazards models were used to identify the impact of LRF ratio on OS. RESULTS: Two hundred and one patients with distal gastroesophageal (48%) or gastric (52%) adenocarcinoma were included. Median age was 65 years, and 146 (73%) were male. On univariate analysis, rectus femoris PET-CT avidity and LRF ratio were significantly associated with overall survival (p < 0.05). LRF ratio was significantly higher in males, early-stage cancer, patients with an ECOG 0 or 1 performance status, patients with albumin > 3.5 mg/dL, and those with moderately differentiated tumor histology. In multivariable regression, gastric cancer stage, albumin, and LRF ratio were significant independent predictors of overall survival (LRF ratio HR = 0.73 (0.56-0.96); p = 0.024). Survival curves showed that the prognostic impact of LRF was associated with metastatic gastric cancer (p = 0.009). CONCLUSIONS: Elevated LRF ratio, a patient-specific PET-CT-based metabolic parameter, was independently associated with an improvement in OS in patients with metastatic gastric cancer. With prospective validation, LRF ratio may be a useful, host-specific metabolic parameter for prognostication in gastric cancer.

The African Research Group for Oncology: A decade fostering colorectal cancer research in Nigeria.

The African Research Group for Oncology (ARGO) was formed in 2013 to undertake methodologically rigorous cancer research in Nigeria, and to strengthen cancer research capacity in the country through training and mentorship of physicians, scientists, and other healthcare workers. Here, we describe how ARGO's work in colorectal cancer (CRC) has evolved over the past decade. This includes the consortium's scientific contributions to the understanding of CRC in Nigeria and globally and its research capacity-building program.

An interpretable AI model for recurrence prediction after surgery in gastrointestinal stromal tumour: an observational cohort study.

Background: There are several models that predict the risk of recurrence following resection of localised, primary gastrointestinal stromal tumour (GIST). However, assessment of calibration is not always feasible and when performed, calibration of current GIST models appears to be suboptimal. We aimed to develop a prognostic model to predict the recurrence of GIST after surgery with both good discrimination and calibration by uncovering and harnessing the non-linear relationships among variables that predict recurrence. Methods: In this observational cohort study, the data of 395 adult patients who underwent complete resection (R0 or R1) of a localised, primary GIST in the pre-imatinib era at Memorial Sloan Kettering Cancer Center (NY, USA) (recruited 1982-2001) and a European consortium (Spanish Group for Research in Sarcomas, 80 sites) (recruited 1987-2011) were used to train an interpretable Artificial Intelligence (AI)-based model called Optimal Classification Trees (OCT). The OCT predicted the probability of recurrence after surgery by capturing non-linear relationships among predictors of recurrence. The data of an additional 596 patients from another European consortium (Polish Clinical GIST Registry, 7 sites) (recruited 1981-2013) who were also treated in the pre-imatinib era were used to externally validate the OCT predictions with regard to discrimination (Harrell's C-index and Brier score) and calibration (calibration curve, Brier score, and Hosmer-Lemeshow test). The calibration of the Memorial Sloan Kettering (MSK) GIST nomogram was used as a comparative gold standard. We also evaluated the clinical utility of the OCT and the MSK nomogram by performing a Decision Curve Analysis (DCA). Findings: The internal cohort included 395 patients (median [IQR] age, 63 [54-71] years; 214 men [54.2%]) and the external cohort included 556 patients (median [IQR] age, 60 [52-68] years; 308 men [55.4%]). The Harrell's C-index of the OCT in the external validation cohort was greater than that of the MSK nomogram (0.805 (95% CI: 0.803-0.808) vs 0.788 (95% CI: 0.786-0.791), respectively). In the external validation cohort, the slope and intercept of the calibration curve of the main OCT were 1.041 and 0.038, respectively. In comparison, the slope and intercept of the calibration curve for the MSK nomogram was 0.681 and 0.032, respectively. The MSK nomogram overestimated the recurrence risk throughout the entire calibration curve. Of note, the Brier score was lower for the OCT compared to the MSK nomogram (0.147 vs 0.564, respectively), and the Hosmer-Lemeshow test was insignificant (P = 0.087) for the OCT model but significant (P < 0.001) for the MSK nomogram. Both results confirmed the superior discrimination and calibration of the OCT over the MSK nomogram. A decision curve analysis showed that the AI-based OCT model allowed for superior decision making compared to the MSK nomogram for both patients with 25-50% recurrence risk as well as those with >50% risk of recurrence. Interpretation: We present the first prognostic models of recurrence risk in GIST that demonstrate excellent discrimination, calibration, and clinical utility on external validation. Additional studies for further validation are warranted. With further validation, these tools could potentially improve patient counseling and selection for adjuvant therapy. Funding: The NCI SPORE in Soft Tissue Sarcoma and NCI Cancer Center Support Grants.

Unique Genomic Alterations and Microbial Profiles Identified in Patients With Gastric Cancer of African, European, and Asian Ancestry: A Novel Path for Precision Oncology.

OBJECTIVE: Here, we characterize differences in the genetic and microbial profiles of GC in patients of African (AFR), European, and Asian ancestry. BACKGROUND: Gastric cancer (GC) is a heterogeneous disease with clinicopathologic variations due to a complex interplay of environmental and biological factors, which may affect disparities in oncologic outcomes.. METHODS: We identified 1042 patients with GC with next-generation sequencing data from an institutional Integrated Mutation Profiling of Actionable Cancer Targets assay and the Cancer Genomic Atlas group. Genetic ancestry was inferred from markers captured by the Integrated Mutation Profiling of Actionable Cancer Targets and the Cancer Genomic Atlas whole exome sequencing panels. Tumor microbial profiles were inferred from sequencing data using a validated microbiome bioinformatics pipeline. Genomic alterations and microbial profiles were compared among patients with GC of different ancestries. RESULTS: We assessed 8023 genomic alterations. The most frequently altered genes were TP53 , ARID1A , KRAS , ERBB2 , and CDH1 . Patients of AFR ancestry had a significantly higher rate of CCNE1 alterations and a lower rate of KRAS alterations ( P < 0.05), and patients of East Asian ancestry had a significantly lower rate of PI3K pathway alterations ( P < 0.05) compared with other ancestries. Microbial diversity and enrichment did not differ significantly across ancestry groups ( P > 0.05). CONCLUSIONS: Distinct patterns of genomic alterations and variations in microbial profiles were identified in patients with GC of AFR, European, and Asian ancestry. Our findings of variation in the prevalence of clinically actionable tumor alterations among ancestry groups suggest that precision medicine can mitigate oncologic disparities.

Histology-Specific Prognostication for Radiation-Associated Soft Tissue Sarcoma.

PURPOSE: Radiation-associated sarcomas (RAS) are rare but aggressive malignancies. We sought to characterize the histology-specific presentation and behavior of soft tissue RAS to improve individualized prognostication. METHODS: A single-institutional prospectively maintained database was queried for all patients with primary, nonmetastatic RAS treated with surgical resection from 1982 to 2019. Patients presenting with the five most common RAS histologies were propensity-matched to those with sporadic tumors of the same histology. Incidence of disease-specific death (DSD) was modeled using cumulative incidence analyses. RESULTS: Among 259 patients with RAS, the five most common histologies were malignant peripheral nerve sheath tumor (MPNST; n = 19), myxofibrosarcoma (n = 20), leiomyosarcoma (n = 24), undifferentiated pleomorphic sarcoma (UPS; n = 55), and angiosarcoma (AS; n = 62). DSD varied significantly by histology (P = .002), with RAS MPNST and UPS having the highest DSD. In unadjusted analysis, RAS MPNST was associated with increased DSD compared with sporadic MPNST (75% v 38% 5-year DSD, P = .002), as was RAS UPS compared with sporadic UPS (49% v 28% 5-year DSD, P = .004). Unadjusted DSD was similar among patients with RAS AS, leiomyosarcoma, or myxofibrosarcoma and sporadic sarcoma of the same histology. After matching RAS to sporadic patients within each histology, DSD only differed between RAS and sporadic MPNST (83% v 46% 5-year DSD, P = .013). Patients with RAS AS presented in such a distinct manner to those with sporadic AS that a successful match was not possible. CONCLUSION: The aggressive presentation of RAS is histology-specific, and DSD is driven by RAS MPNST and UPS histologies. Despite the aggressive presentation, standard prognostic factors can be used to estimate risk of DSD among most RAS. In MPNST, radiation association should be considered to independently associate with markedly higher risk of DSD.

Fifty years of pancreas cancer care.

Resulting from 50 years of innovation, operations for pancreatic neoplasms can now be performed safely, albeit with significant but manageable morbidity. Molecular diagnosis has allowed for the identification of multiple distinct histopathologies with variable natural histories. Observation is now a strategy for selected indolent cysts and some neuroendocrine neoplasms. For ductal pancreatic adenocarcinoma, a long-term cure remains elusive and will require more than surgical resection for meaningful progress.

Five decades of sarcoma care at Memorial Sloan Kettering Cancer Center.

Early studies of the management of soft tissue sarcoma at Memorial Sloan Kettering Cancer Center were influenced by development of robust prospective long-term databases. Increasing capacity for molecular diagnostics has identified a myriad of subtypes with definable natural history. Accurate identification of tissue-specific risk of recurrence and disease-specific survival have increasingly allowed selective use of surgery, radiation therapy, and target-specific cytotoxic and immune therapies.

Validation of the Memorial Sloan Kettering Gastric Cancer Post-Resection Survival Nomogram: Does It Stand the Test of Time?

BACKGROUND: The Memorial Sloan Kettering Cancer Center (MSK) nomogram combined both gastroesophageal junction (GEJ) and gastric cancer patients and was created in an era from patients who generally did not receive neoadjuvant chemotherapy. We sought to reevaluate the MSK nomogram in the era of multidisciplinary treatment for GEJ and gastric cancer. STUDY DESIGN: Using data on patients who underwent R0 resection for GEJ or gastric cancer between 2002 and 2016, the C-index of prediction for disease-specific survival (DSS) was compared between the MSK nomogram and the American Joint Committee on Cancer (AJCC) 8th edition staging system after segregating patients by tumor location (GEJ or gastric cancer) and neoadjuvant treatment. A new nomogram was created for the group for which both systems poorly predicted prognosis. RESULTS: During the study period, 886 patients (645 gastric and 241 GEJ cancer) underwent up-front surgery, and 999 patients (323 gastric and 676 GEJ) received neoadjuvant treatment. Compared with the AJCC staging system, the MSK nomogram demonstrated a comparable C-index in gastric cancer patients undergoing up-front surgery (0.786 vs 0.753) and a better C-index in gastric cancer patients receiving neoadjuvant treatment (0.796 vs 0.698). In GEJ cancer patients receiving neoadjuvant chemotherapy, neither the MSK nomogram nor the AJCC staging system performed well (C-indices 0.647 and 0.646). A new GEJ nomogram was created based on multivariable Cox regression analysis and was validated with a C-index of 0.718. CONCLUSIONS: The MSK gastric cancer nomogram's predictive accuracy remains high. We developed a new GEJ nomogram that can effectively predict DSS in patients receiving neoadjuvant treatment.

Clinical genomic profiling in the management of patients with soft tissue and bone sarcoma.

There are more than 70 distinct sarcomas, and this diversity complicates the development of precision-based therapeutics for these cancers. Prospective comprehensive genomic profiling could overcome this challenge by providing insight into sarcomas' molecular drivers. Through targeted panel sequencing of 7494 sarcomas representing 44 histologies, we identify highly recurrent and type-specific alterations that aid in diagnosis and treatment decisions. Sequencing could lead to refinement or reassignment of 10.5% of diagnoses. Nearly one-third of patients (31.7%) harbor potentially actionable alterations, including a significant proportion (2.6%) with kinase gene rearrangements; 3.9% have a tumor mutational burden ≥10 mut/Mb. We describe low frequencies of microsatellite instability (<0.3%) and a high degree of genome-wide loss of heterozygosity (15%) across sarcomas, which are not readily explained by homologous recombination deficiency (observed in 2.5% of cases). In a clinically annotated subset of 118 patients, we validate actionable genetic events as therapeutic targets. Collectively, our findings reveal the genetic landscape of human sarcomas, which may inform future development of therapeutics and improve clinical outcomes for patients with these rare cancers.

Surgical resection for intraductal papillary mucinous neoplasm in the older population.

BACKGROUND: Surgery for intraductal papillary mucinous neoplasm (IPMN) in older adults requires a careful balance of risk and benefit. We sought to analyze patient outcomes in the older individuals after pancreatic resection for IPMN. METHODS: Retrospective analysis of a prospectively maintained database was performed for patients 65 years or older undergoing IPMN resection between January 1, 2012 and December 31, 2017. Statistical analysis was performed based on age and Memorial Sloan Kettering Frailty Index (MSKFI) score. RESULTS: 148 patients underwent resection of an IPMN, including five patients who required two operations for recurrent disease. Median age at surgery was 74 (range, 65-90 years), and 52% were male. Most patients underwent pancreaticoduodenectomy (53%) or distal pancreatectomy/splenectomy (35%). An associated adenocarcinoma was seen on pathology for 56 patients (37%). Median hospital length of stay was 7 days (range, 4-46 days). Grade 3 or higher post-operative complications on the Clavien-Dindo classification scale were seen in 20%. No patient died within 30-days. Patient outcomes were evaluated by age, split at age ≥75 (considered "elderly"), and separately by MSKFI score. No differences in post-operative morbidity or mortality was seen when stratified by age (65 - 74 vs > 75 years) or by MSKFI frailty score. CONCLUSION: Pancreatic resection can be safely performed in selected patients 65 years and older with low morbidity and mortality. More analysis is needed to determine if MSKFI score is a useful predictor of complications in older individuals.

Association of Obesity with Worse Operative and Oncologic Outcomes for Patients Undergoing Gastric Cancer Resection.

BACKGROUND: How obesity has an impact on operative and oncologic outcomes for gastric cancer patients is unclear, and the influence of obesity on response to neoadjuvant chemotherapy (NAC) has not been evaluated. METHODS: Patients who underwent curative gastrectomy for primary gastric cancer between 2000 and 2018 were retrospectively identified. After stratification for NAC, operative morbidity, mortality, overall survival (OS), and disease-specific survival (DSS) were compared among three body mass index (BMI) categories: normal BMI (< 25 kg/m2), mild obesity (25-35 kg/m2), and severe obesity (≥ 35 kg/m2). RESULTS: During the study period, 984 patients underwent upfront surgery, and 484 patients received NAC. Tumor stage did not differ among the BMI groups. However, the rates of pathologic response to NAC were significantly lower for the patients with severe obesity (10% vs 40%; p < 0.001). Overall complications were more frequent among the obese patients (44.3% for obese vs 24.9% for normal BMI, p < 0.001). Intraabdominal infections were also more frequent in obese patients (13.9% for obese vs 4.7% for normal BMI, p = 0.001). In the upfront surgery cohort, according to the BMI, OS and DSS did not differ, whereas in the NAC cohort, severe obesity was independently associated with worse OS [hazard ratio (HR) 1.87; 95% confidence interval (CI) 1.01-3.48; p = 0.047] and disease-specific survival (DSS) (HR 2.08; 95% CI 1.07-4.05; p = 0.031). CONCLUSION: For the gastric cancer patients undergoing curative gastrectomy, obesity was associated with significantly lower rates of pathologic response to NAC and more postoperative complications, as well as shorter OS and DSS for the patients receiving NAC.

Prophylactic Lateral Neck Dissection for Medullary Thyroid Carcinoma is not Associated with Improved Survival.

BACKGROUND: Patients with medullary thyroid carcinoma (MTC) often receive lateral lymph node dissection with total thyroidectomy when calcitonin levels are elevated, even in the absence of structural disease, but the effect of this intervention on disease-specific outcomes is not known. PATIENTS AND METHODS: We retrospectively reviewed patients from 1986 to 2017 who underwent thyroidectomy with curative intent for MTC at our institution. The association of disease-specific survival and clinicopathologic features was examined using univariate and multivariate Cox regression. RESULTS: We identified 316 patients who underwent curative resection for MTC. Overall and disease-specific survival were 76% and 86%, respectively, at 10 years. To investigate the effect of prophylactic ipsilateral lateral lymph node dissection, we analyzed 89 patients without known structural disease in the neck lymph nodes at the time of resection and preoperative calcitonin > 200 pg/ml, of whom 45 had an ipsilateral lateral lymph node dissection (LND) and 44 did not. There were no differences in tumor size or preoperative calcitonin levels. There was no difference at 10 years in cumulative incidence of recurrence in the neck (20.9% LND vs. 30.4% no LND, p = 0.46), cumulative incidence of distant recurrence (18.3% vs. 18.4%, p = 0.97), disease-specific survival (86% vs. 93%, p = 0.53), or overall survival (82% vs. 90%, p = 0.6). CONCLUSION: Lateral neck dissection in the absence of clinical or radiologic abnormal lymph nodes is not associated with improved survival in patients with MTC.

Outcomes of Neoadjuvant Chemotherapy for Clinical Stages 2 and 3 Gastric Cancer Patients: Analysis of Timing and Site of Recurrence.

BACKGROUND: This study aimed to analyze timing and sites of recurrence for patients receiving neoadjuvant chemotherapy for gastric cancer. Neoadjuvant chemotherapy followed by surgical resection is the standard treatment for locally advanced gastric cancer in the West, but limited information exists as to timing and patterns of recurrence in this setting. METHODS: Patients with clinical stage 2 or 3 gastric cancer treated with neoadjuvant chemotherapy followed by curative-intent resection between January 2000 and December 2015 were analyzed for 5-year recurrence-free survival (RFS) as well as timing and site of recurrence. RESULTS: Among 312 identified patients, 121 (38.8%) experienced recurrence during a median follow-up period of 46 months. The overall 5-year RFS rate was 58.9%, with RFS rates of 95.8% for ypT0N0, 81% for ypStage 1, 77.4% for ypStage 2, and 22.9% for ypStage 3. The first site of recurrence was peritoneal for 49.6%, distant (not peritoneal) for 45.5%, and locoregional for 11.6% of the patients. The majority of the recurrences (84.3%) occurred within 2 years. Multivariate analysis showed that ypT4 status was an independent predictor for recurrence within 1 year after surgery (odds ratio, 2.58; 95% confidence interval, 1.10-6.08; p = 0.030). CONCLUSIONS: The majority of the recurrences for patients with clinical stage 2 or 3 gastric cancer who received neoadjuvant chemotherapy and underwent curative resection occurred within 2 years. After neoadjuvant chemotherapy, pathologic T stage was a useful risk predictor for early recurrence.