1
|
Sibia US, Millen JC, Klune JR, Bilchik A, Foshag LJ. Analysis of 10-year trends in Medicare Physician Fee Schedule payments in surgery. Surgery 2024; 175:920-926. [PMID: 38262816 DOI: 10.1016/j.surg.2023.12.012] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/12/2023] [Revised: 11/29/2023] [Accepted: 12/11/2023] [Indexed: 01/25/2024]
Abstract
BACKGROUND Medicare expenditures have steadily increased over the decades, and yet Medicare Physician Fee Schedule payments for individual services have declined. We examine trends in Medicare Physician Fee Schedule payments for office visits, inpatient visits, and surgical procedures. METHODS The Medicare Physician Fee Schedule Look-Up Tool was queried for payment data for office visits, inpatient visits, and surgical procedures between 2013 and 2023. All data were adjusted for inflation using the Consumer Price Index. Trends in payments were calculated for 5 common procedures in each surgical specialty. Trends in aggregate national health expenditures were compared to Medicare Physician Fee Schedule payments for physician services from 2013 to 2021. RESULTS The Consumer Price Index increased by 29.3% from 2013 to 2023. Inflation-adjusted per-visit Medicare Physician Fee Schedule payments decreased by 12.2% for outpatient office visits, 19.1% for inpatient visits, and 22.8% for surgical procedures from 2013 to 2023. This varied by surgical specialty: vascular (-25.8%), endocrine (-22.0%), general surgery (-27.0%), thoracic (-19.2%), surgical oncology (-22.1%), breast (-22.4%), urology (-2.2%), neurosurgery (-22.8%), obstetrics/gynecology (-19.9%), and orthopedics (-24.7%). Adjusted for inflation, national health expenditures increased by 33.9% for physician services from 2013 to 2021. In comparison, Medicare Physician Fee Schedule payments over the same time period 2013 to 2021 increased by 1.3% for outpatient office visits but decreased by 10.6% for inpatient visits and 9.8% for surgical procedures. CONCLUSION Controlling rising national health expenditures is important and necessary, but 10 years of declining Medicare Physician Fee Schedule payments on a per-procedure basis in surgery would suggest that this strategy alone may not achieve those goals and could ultimately threaten access to quality surgical care. Surgeons must advocate for permanent payment reforms.
Collapse
|
2
|
Sibia US, Klune JR, Saiolghalam S, Bilchik A. Early Experiences With Bundled Payments for Care Improvement for Major Bowel Surgery. Am Surg 2024:31348241241618. [PMID: 38523411 DOI: 10.1177/00031348241241618] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 03/26/2024]
Abstract
BACKGROUND Bundled Payment (BP) models are becoming more common in surgery. We share our early experiences with Bundled Payments for Care Improvement for major bowel surgery. METHODS Patients undergoing major bowel surgery between January and October 2021 were identified using Medicare Severity-Diagnosis Related Group (MS-DRG) codes. Major drivers of cost in a BP model are reported and compared to the Fee-For-Service (FFS) payment model. RESULTS A total of 202 cases (173 FFS vs 29 BP) were analyzed. The mean BP cost per Clinical Episode was $28,340. Eleven patients (38%) in the BP model had costs greater than the Target Price. The drivers of cost in the BP model were 59% acute care facility, 17% physician services, 9% post-acute care facilities, 8% other, and 7% readmissions. Clinical Episode of care costs varied considerably by MS-DRG case complexity. Robotic surgery increased costs by 35% (mean increase $3724, P < .01). The 90-day readmission rate was 17% for a mean cost of $11,332 per readmission. Three patients (10%) were discharged to a skilled nursing facility at an average cost of $11,009, while fifteen patients (52%) received home health services at a mean cost of $2947. Acute care facility costs were similar in the BP vs FFS groups (mean difference $1333, P = .22). CONCLUSIONS Patients undergoing major bowel surgery are a heterogeneous population. Physicians are ideally positioned to deliver high-value, patient-centered care and are crucial to the success of a BP model. The post-acute care setting is a key component of improving efficiency and quality of care.
Collapse
|
3
|
Christopher W, Nassoiy S, Marcus R, Keller J, Chang SC, Fischer T, Bilchik A, Goldfarb M. Neoadjuvant chemotherapy improves outcomes in resectable pancreatic adenocarcinoma. SURGERY IN PRACTICE AND SCIENCE 2022; 11:100136. [PMID: 39845164 PMCID: PMC11749821 DOI: 10.1016/j.sipas.2022.100136] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/12/2022] [Accepted: 09/26/2022] [Indexed: 11/23/2022] Open
Abstract
Background/Objectives Systemic chemotherapy is recommended for all stages of pancreatic ductal adenocarcinoma (PDAC), with a recent shift towards neoadjuvant chemotherapy (NAC) for resectable PDAC. The objective of this study was to compare outcomes of NAC versus AC for early stage resectable PDAC in the NCDB. Methods: Patients aged 18 or older with stage I or II PDAC in the National Cancer Database (NCDB) from 2010 to 2017 were identified. Logistic regression evaluated oncologic outcomes. Kaplan-Meier method followed by Cox proportional-hazards regression with inverse probability of treatment weighting (IPTW) using propensity score matching was used to compare overall survival (OS). Results NAC led to a 13% risk reduction of a positive resection margin (OR:0.87; 95%CI 0.78-0.97), and a 59% decreased risk of positive lymph nodes (OR:0.41; 95%CI 0.38-0.45). The median OS for all patients treated with NAC was 2.56 years versus 1.95 years for surgery +/- AC (p< 0.001). NAC had an OS benefit for all patients (HR:0.80; 95%CI 0.78-0.83), as well as for Stage I (HR:0.89; 95%CI 0.84-0.94) and Stage II patients (HR:0.71; 95%CI 0.68-0.75). Conclusions NAC appears to be associated with a survival benefit for patients with resectable PDAC. NAC also decreased the risk of a positive resection margin and positive lymph nodes at the time of surgery.
Collapse
|
4
|
Christopher W, Nassoiy S, Marcus R, Keller J, Chang SC, Fischer T, Bilchik A, Goldfarb M. Prognostic indicators for undifferentiated carcinoma with/without osteoclast-like giant cells of the pancreas. HPB (Oxford) 2022; 24:1757-1769. [PMID: 35780038 DOI: 10.1016/j.hpb.2022.05.1344] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 02/03/2022] [Revised: 05/03/2022] [Accepted: 05/26/2022] [Indexed: 12/12/2022]
Abstract
BACKGROUND Undifferentiated carcinoma of the pancreas (UPC) is a rare malignancy. There are no standardized guidelines for treatment. Current management has been extrapolated from smaller reviews. METHODS 858 patients with UPC were identified in the 2004-2017 NCDB. Kaplan-Meier method followed by Cox proportional-hazards regression examined independent prognostic factors associated with overall survival (OS). Logistic regression analyses were performed to determine independent predictors of surgical intervention and the status of surgical resection by histologic subtype. RESULTS Patients with osteoclast-like giant cells (OCLGC) had a longer median OS compared to those without (aHR 0.52: 95% CI 0.41-0.67). Of the non-OCLGC subtypes, pleomorphic large cell demonstrated the shortest median OS (2.4 months). Surgical resection was associated with improved survival in all histologies except for pleomorphic cell carcinoma. R0 resection and negative lymph nodes were independently associated with an improved OS. CONCLUSION This is the largest database review published to date on UCP. OCLGC histology is associated with an improved survival compared to those without OCLGC. Of the non-OCLGC subtypes, pleomorphic large cell is associated with the shortest overall survival. Surgical resection is associated with a significant survival advantage for all histologies except for pleomorphic cell carcinoma.
Collapse
|
5
|
Steele SR, Bilchik A, Johnson EK, Nissan A, Peoples GE, Eberhardt JS, Kalina P, Petersen B, BrüCher B, Protic M, Avital I, Stojadinovic A. Time-dependent Estimates of Recurrence and Survival in Colon Cancer: Clinical Decision Support System Tool Development for Adjuvant Therapy and Oncological Outcome Assessment. Am Surg 2020. [DOI: 10.1177/000313481408000514] [Citation(s) in RCA: 9] [Impact Index Per Article: 1.8] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/17/2022]
Abstract
Unanswered questions remain in determining which high-risk node-negative colon cancer (CC) cohorts benefit from adjuvant therapy and how it may differ in an equal access population. Machine-learned Bayesian Belief Networks (ml-BBNs) accurately estimate outcomes in CC, providing clinicians with Clinical Decision Support System (CDSS) tools to facilitate treatment planning. We evaluated ml-BBNs ability to estimate survival and recurrence in CC. We performed a retrospective analysis of registry data of patients with CC to train–test–crossvalidate ml-BBNs using the Department of Defense Automated Central Tumor Registry (January 1993 to December 2004). Cases with events or follow-up that passed quality control were stratified into 1-, 2-, 3-, and 5-year survival cohorts. ml-BBNs were trained using machine-learning algorithms and k-fold crossvalidation and receiver operating characteristic curve analysis used for validation. BBNs were comprised of 5301 patients and areas under the curve ranged from 0.85 to 0.90. Positive predictive values for recurrence and mortality ranged from 78 to 84 per cent and negative predictive values from 74 to 90 per cent by survival cohort. In the 12-month model alone, 1,132,462,080 unique rule sets allow physicians to predict individual recurrence/mortality estimates. Patients with Stage II (N0M0) CC benefit from chemotherapy at different rates. At one year, all patients older than 73 years of age with T2–4 tumors and abnormal carcinoembryonic antigen levels benefited, whereas at five years, all had relative reduction in mortality with the largest benefit amongst elderly, highest T-stage patients. ml-BBN can readily predict which high-risk patients benefit from adjuvant therapy. CDSS tools yield individualized, clinically relevant estimates of outcomes to assist clinicians in treatment planning.
Collapse
|
6
|
Ito Y, Miyashiro I, Ishikawa T, Akazawa K, Fukui K, Katai H, Nunobe S, Oda I, Isobe Y, Tsujitani S, Ono H, Tanabe S, Fukagawa T, Suzuki S, Kakeji Y, Sasako M, Bilchik A, Fujita M. Determinant Factors on Differences in Survival for Gastric Cancer Between the United States and Japan Using Nationwide Databases. J Epidemiol 2020; 31:241-248. [PMID: 32281553 PMCID: PMC7940976 DOI: 10.2188/jea.je20190351] [Citation(s) in RCA: 24] [Impact Index Per Article: 4.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/28/2022] Open
Abstract
Background Although the incidence and mortality have decreased, gastric cancer (GC) is still a public health issue globally. An international study reported higher survival in Korea and Japan than other countries, including the United States. We examined the determinant factors of the high survival in Japan compared with the United States. Methods We analysed data on 78,648 cases from the nationwide GC registration project, the Japanese Gastric Cancer Association (JGCA), from 2004–2007 and compared them with 16,722 cases from the Surveillance, Epidemiology, and End Results Program (SEER), a United States population-based cancer registry data from 2004–2010. We estimated 5-year relative survival and applied a multivariate excess hazard model to compare the two countries, considering the effect of number of lymph nodes (LNs) examined. Results Five-year relative survival in Japan was 81.0%, compared with 45.0% in the United States. After controlling for confounding factors, we still observed significantly higher survival in Japan. Among N2 patients, a higher number of LNs examined showed better survival in both countries. Among N3 patients, the relationship between number of LNs examined and differences in survival between the two countries disappeared. Conclusion Although the wide differences in GC survival between Japan and United States can be largely explained by differences in the stage at diagnosis, the number of LNs examined may also help to explain the gaps between two countries, which is related to stage migration.
Collapse
|
7
|
Santamaria-Barria JA, Bilchik A. ASO Author Reflections: The Changing Landscape for Industry Support to Surgical Oncologists. Ann Surg Oncol 2019; 26:828-829. [PMID: 31691115 DOI: 10.1245/s10434-019-07917-4] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/19/2019] [Indexed: 11/18/2022]
|
8
|
Santamaria-Barria JA, Stern S, Khader A, Garland-Kledzik M, Scholer AJ, Fischer T, Bilchik A. Changing Trends in Industry Funding for Surgical Oncologists. Ann Surg Oncol 2019; 26:2327-2335. [DOI: 10.1245/s10434-019-07380-1] [Citation(s) in RCA: 8] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/16/2019] [Indexed: 12/21/2022]
|
9
|
Protic M, Bilchik A, Nissan A, Knezevic SU, Kukic B, Kresoja M, Veljkovic R, Zivojinov M, Stojadinovic A. International prospective multi-center clinical trial with adherence to surgical and pathological quality measures: Influence of body mass index (BMI) on outcome in colon cancer. Ann Oncol 2018. [DOI: 10.1093/annonc/mdy151.241] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/12/2022] Open
|
10
|
Teng A, Nguyen T, Bilchik A, O'Connor V, Lee DY. Implications of prolonged time to pancreaticoduodenectomy after neoadjuvant chemoradiation: Analysis of the National Cancer Database. J Clin Oncol 2018. [DOI: 10.1200/jco.2018.36.4_suppl.473] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.1] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/20/2022] Open
Abstract
473 Background: For patients with pancreatic adenocarcinoma (PA), the optimal time interval between neoadjuvant chemoradiation (CR) to surgical resection has not been well established. The National Cancer Database (NCDB) was used to evaluate the impact of radiation-surgery (RS) interval on outcomes. Methods: The NCDB from 2006-2014 was queried for patients ≥18 years old diagnosed with PA who received CR prior to surgery. Survival and short-term outcomes were compared between patients who had a Whipple procedure performed ≤12 weeks and > 12 weeks after completion of CR therapy. Results: 1610 patients met selection criteria. Average RS interval was 58.2 ± 39.5 days. 1419 patients had RS interval ≤12 weeks (mean 47.4 days) and 191 had RS interval > 12 weeks (mean 138.8 days). Age, race, gender, income, type of treatment facility, CA 19-9 levels, types of chemotherapy and radiation dosage administered were similar between the two groups. Mean tumor size was 32.2 mm in the ≤12 week group and 34.9 mm in the > 12 week group (p = 0.021). There was a higher proportion of patients with clinical stage III cancers in the > 12 weeks group than in the ≤12 weeks group (33.5% vs 14%). Short-term morbidity and mortality was not significantly different between the two groups in terms of length of stay, readmission within 30 days, 30-day and 90-day mortality. However, a long-term survival benefit was observed in the > 12 week group (median 25.8 months in ≤12 weeks vs 30.2 months in > 12 weeks, p = 0.049) that appears to persist. An interval > 12 weeks was associated with significantly prolonged survival on multivariate analysis (HR 0.80 (0.65-0.99 95% CI, p = 0.042)). Higher clinical stage and positive surgical margins were independently associated with worse survival. Conclusions: Surgical resection beyond 12 weeks after CR for PA did not worsen surgical outcomes. Waiting may contribute to better patient selection, especially those with larger tumors and higher clinical stage. In the absence of progressive disease, patients need to be continuously evaluated for surgical resection after CR.
Collapse
|
11
|
Vuong B, Dehal A, Graff-Baker A, Chang SC, Foshag LJ, Bilchik A, Goldfarb M. Effect of palliative surgery, chemotherapy, and radiation in stage IV pancreatic cancer. J Clin Oncol 2017. [DOI: 10.1200/jco.2017.35.31_suppl.233] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/20/2022] Open
Abstract
233 Background: Despite significant advances in multi-modality treatment for pancreatic adenocarcinoma (PC), prognosis for stage IV PC remains poor. While reducing suffering and optimizing quality of life are the primary goals of palliative therapies, these interventions may extend overall survival. We examined the impact on survival when aggressive palliative treatments including surgery, chemotherapy, or radiation were employed in end-of-life care. Methods: The 2004-2014 National Cancer Data Base (NCDB) was queried to identify patients with stage IV PC that did not undergo primary surgical resection. Univariate (Kaplan Meier and log-rank) and multivariable (Cox proportional hazard) analyses were used to assess the associations between patient characteristics, use of palliative therapies, and overall survival. Results: Of 72,736 patients identified with metastatic stage IV PC, 2,097 (3%) underwent surgical palliation (ST), 5,615 (8%) received palliative chemotherapy (CT), 940 (1%) received palliative radiation (RT), 1,163 (2%) received multimodality treatment (MMT), and 62,921 (87%) had no aggressive palliative intervention (NT). The choice of palliative therapy, if any, was influenced by all demographic and tumor variables except for gender (all p < 0.001). Median OS was greatest after CT (5.09 months, p < 0.001) compared to any other modality (NT: 3.45months, ST: 3.71months, RT: 3.25months, MMT: 4.47months). This remained true regardless of age, gender, race/ethnicity, insurance, and facility type. After adjusting for all demographic and tumor factors, use of CT decreased the annual risk of death by 20% (HR = 0.8; 95%CI [0.77, 0.82]) and MMT by 10% (HR = 0.9; 95%CI [0.84, 0.96]). Employment of RT increased risk of death by 9% (HR = 1.09; 95%CI [1.01, 1.17]) and ST did not affect OS (HR = 1.01; 95%CI [0.96,1.06]). Conclusions: Despite advances in palliative treatments, Stage IV PC arries a dismal prognosis. Palliative RT may shorten survival. Equivalent survival for ST versus NT suggests that this may be beneficial in the appropriate patient. Palliative CT independently improved survival by approximately 6 weeks and should be considered in patients that want to extend survival and can tolerate the toxicity.
Collapse
|
12
|
Vuong B, Dehal A, Graff-Baker AN, Chang SC, Foshag L, Bilchik A, Goldfarb MR. Effect of palliative surgery, chemotherapy, and radiation in stage IV pancreatic cancer. J Clin Oncol 2017. [DOI: 10.1200/jco.2017.35.15_suppl.e15707] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/20/2022] Open
Abstract
e15707 Background: Despite significant advances in multi-modality treatment for pancreatic adenocarcinoma (PC), prognosis for stage IV PC remains poor. While reducing suffering and optimizing quality of life are the primary goals of palliative therapies, these interventions may extend overall survival. We examined the impact on survival when aggressive palliative treatments including surgery, chemotherapy, or radiation were employed in end-of-life care. Methods: The 2004-2014 National Cancer Data Base (NCDB) was queried to identify patients with stage IV PC that did not undergo primary surgical resection. Univariate (Kaplan Meier and log-rank) and multivariable (Cox proportional hazard) analyses were used to assess the associations between patient characteristics, use of palliative therapies, and overall survival. Results: Of 72,736 patients identified with metastatic stage IV PC, 2,097 (3%) underwent surgical palliation (ST), 5,615 (8%) received palliative chemotherapy (CT), 940 (1%) received palliative radiation (RT), 1,163 (2%) received multimodality treatment (MMT), and 62,921 (87%) had no aggressive palliative intervention (NT). The choice of palliative therapy, if any, was influenced by all demographic and tumor variables except for gender (all p < 0.001). Median OS was greatest after CT (5.09 months, p < 0.001) compared to any other modality (NT: 3.45months, ST: 3.71months, RT: 3.25months, MMT: 4.47months). This remained true regardless of age, gender, race/ethnicity, insurance, and facility type. After adjusting for all demographic and tumor factors, use of CT decreased the annual risk of death by 20% (HR = 0.8; 95%CI [0.77, 0.82]) and MMT by 10% (HR = 0.9; 95%CI [0.84, 0.96]). Employment of RT increased risk of death by 9% (HR = 1.09; 95%CI [1.01, 1.17]) and ST did not affect OS (HR = 1.01; 95%CI [0.96,1.06]). Conclusions: Despite advances in palliative treatments, Stage IV PC arries a dismal prognosis. Palliative RT may shorten survival. Equivalent survival for ST versus NT suggests that this may be beneficial in the appropriate patient. Palliative CT independently improved survival by approximately 6 weeks and should be considered in patients that want to extend survival and can tolerate the toxicity.
Collapse
|
13
|
Dehal A, Vuong B, Graff-Baker AN, Lee J, Bilchik A, Goldfarb MR. Effect of colon cancer sidedness compared to tumor biology on stage-specific survival. J Clin Oncol 2017. [DOI: 10.1200/jco.2017.35.4_suppl.543] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.1] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/20/2022] Open
Abstract
543 Background: Recent studies have reported a survival difference between right colon cancer (RCC) and left colon cancer (LCC) but were limited by lack of data on surgical therapy and underlying tumor biology including microsatellite instability (MSI) status. Our objective was to examine the influence of tumor sidedness on stage-specific survival after adjustment for those factors using a large national database. Methods: Adult patients with primary invasive colon cancer were identified from the National Cancer Data Base (NCDB) between 2009 and 2012. Cecum, ascending colon, hepatic flexure, and transverse colon were classified as “RCC” and sigmoid, descending colon, and splenic flexure were classified as “LCC”. Differences in 3-year overall survival (OS) between RCC versus LCC, stratified by stage, were obtained. Results: Of 24,092 patients, 9,095 (37.75%) presented with a LCC and 14,997 (62.25%) with a RCC. After adjustment for confounding factors, including demographics, MSI, KRAS, nodal retrieval and treatment, patients with RCC stage I, III, or IV had a 31-35% decreased OS compared to those with a LCC. However, in patients with stage II disease, RCC had a 15% improvement in OS compared to LCC (Table 1). MSI positivity was not significantly different in RCC versus LCC stage for stage nor did it contribute to decreased OS for either side. Conclusions: Colon cancer sidedness is an independent risk factor for decreased survival after adjusting for potential confounders including tumor biology and treatment characteristics. For patients with stage I, II, and IV disease, RCC portends a decreased survival compared to LCC, whereas the opposite is true for patients with stage II disease. Further research is needed to elucidate the exact the role of sidedness in risk-profiling and staging for colon cancer patients. [Table: see text]
Collapse
|
14
|
Deutsch GB, Kirchoff DD, Bailey M, Vitug S, Flaherty DC, Foshag LJ, Bilchik A, Faries MB. Gastrointestinal metastases from melanoma: Does surgical resection improve survival when performed in combination with immunotherapy? J Clin Oncol 2015. [DOI: 10.1200/jco.2015.33.3_suppl.340] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.1] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/20/2022] Open
Abstract
340 Background: Prior studies have shown the benefits of metastasectomy and more recently, immunotherapy, in Stage IV melanoma. We report the largest series of gastrointestinal (GI) melanoma metastases, in order to clarify the role of surgical resection in the setting of systemic treatment options. Methods: A review of our institutional melanoma database identified all patients with hollow viscus GI metastases diagnosed between 1971 and 2013. Patient age and sex; primary tumor location, thickness and ulceration; GI-metastasis free interval; and types of treatment were analyzed. The primary endpoint was melanoma-specific survival (MSS), measured from the diagnosis of GI metastases. Survival analyses compared combination therapy to immunotherapy (i.e. vaccine, interleukin-2, interferon, ipilimumab) or surgery alone. Operative patients were further stratified by surgical intent (curative or palliative). Results: The 393 study patients had a median GI-metastasis free interval of 49.5 (0 – 416.0) months from the initial diagnosis; their mean age at GI metastases was of 53.5 ± 14.1 years. Median MSS was 20.0 months with surgery plus immunotherapy (n=160), 13.0 months with surgery alone (n=111), 8.0 months with immunotherapy alone (n=64), and 5.0 months with neither treatment (n=58) (p<0.001). By multivariable analysis, age at GI metastasis diagnosis (HR=1.014; p=.014), truncal melanoma site (HR=1.659; p=.017), immunotherapy (HR=0.675; p=.030), and metastasectomy (HR=0.544; p<.001) were significant predictors of MSS. Surgical intent was known in 98 patients: curative surgery had a median MSS of 30.0 months (n=49) and palliative surgery had a median MSS of 12.0 months (n=49) (p=.015). Conclusions: Treatment of GI melanoma metastases should use a combination of metastasectomy and immunotherapy, instead of either modality alone. Select patients are likely to obtain an enhanced benefit from GI metastasectomy, specifically curative surgery, in the setting of newer immunotherapy options.
Collapse
|
15
|
Steele SR, Bilchik A, Johnson EK, Nissan A, Peoples GE, Eberhardt JS, Kalina P, Petersen B, Brücher B, Protic M, Avital I, Stojadinovic A. Time-dependent estimates of recurrence and survival in colon cancer: clinical decision support system tool development for adjuvant therapy and oncological outcome assessment. Am Surg 2014; 80:441-453. [PMID: 24887722] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 06/03/2023]
Abstract
Unanswered questions remain in determining which high-risk node-negative colon cancer (CC) cohorts benefit from adjuvant therapy and how it may differ in an equal access population. Machine-learned Bayesian Belief Networks (ml-BBNs) accurately estimate outcomes in CC, providing clinicians with Clinical Decision Support System (CDSS) tools to facilitate treatment planning. We evaluated ml-BBNs ability to estimate survival and recurrence in CC. We performed a retrospective analysis of registry data of patients with CC to train-test-crossvalidate ml-BBNs using the Department of Defense Automated Central Tumor Registry (January 1993 to December 2004). Cases with events or follow-up that passed quality control were stratified into 1-, 2-, 3-, and 5-year survival cohorts. ml-BBNs were trained using machine-learning algorithms and k-fold crossvalidation and receiver operating characteristic curve analysis used for validation. BBNs were comprised of 5301 patients and areas under the curve ranged from 0.85 to 0.90. Positive predictive values for recurrence and mortality ranged from 78 to 84 per cent and negative predictive values from 74 to 90 per cent by survival cohort. In the 12-month model alone, 1,132,462,080 unique rule sets allow physicians to predict individual recurrence/mortality estimates. Patients with Stage II (N0M0) CC benefit from chemotherapy at different rates. At one year, all patients older than 73 years of age with T2-4 tumors and abnormal carcinoembryonic antigen levels benefited, whereas at five years, all had relative reduction in mortality with the largest benefit amongst elderly, highest T-stage patients. ml-BBN can readily predict which high-risk patients benefit from adjuvant therapy. CDSS tools yield individualized, clinically relevant estimates of outcomes to assist clinicians in treatment planning.
Collapse
|
16
|
Steele S, Bilchik A, Johnson E, Nissan A, Peoples GE, Eberhardt J, Kalina P, Petersen B, Bruecher BLDM, Protic M, Avital I, Stojadinovic A. Time-dependent estimates of recurrence and survival in colon cancer: clinical decision support system tool development for adjuvant therapy and oncological outcome assessment. J Clin Oncol 2013. [DOI: 10.1200/jco.2013.31.15_suppl.e14500] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/20/2022] Open
Abstract
e14500 Background: Unanswered questions remain regarding treatment efficacy in colon cancer (CC), especially those determining high-risk node-negative cohorts that may benefit from adjuvant therapy. We sought to evaluate the use of machine learning and classification modeling to estimate survival and recurrence in CC. Methods: We used the Department of Defense Automated Central Tumor Registry (ACTUR) to identify primary CC patients treated between January 1993 and December 2004. Cases with events or follow-up that passed quality control were stratified into one-, two-, three-, and five-year survival cohorts. ml-BBNs were trained using machine-learning algorithms and k-fold cross-validation, and receiver operating characteristic (ROC) curve analysis used for validation. Results: There were 5,301 cases stratified into cohorts. Survival cohort Areas-Under-the-Curve (AUCs) ranged from 0.85–0.90, positive-predictive-values (PPVs) for recurrence and mortality ranged from 78-84% and negative-predictive-values (NPVs) from 74-90%. Cross-validation showed that the ml-BBNs produce robust individual estimates of recurrence (p<0.001) and mortality (p<0.001) based on readily available clinical-pathological information in the context of adjuvant chemotherapy. Conclusions: Tumor registry data and machine-learned Bayesian Belief Networks produce robust classifiers. These Clinical Decision Support System tools yield clinically relevant estimates of outcomes that may assist clinicians in treatment planning.
Collapse
|
17
|
Mazeh H, Mizrahi I, Ilyayev N, Halle D, Brücher B, Bilchik A, Protic M, Daumer M, Stojadinovic A, Itzhak A, Nissan A. The Diagnostic and Prognostic Role of microRNA in Colorectal Cancer - a Comprehensive review. J Cancer 2013; 4:281-95. [PMID: 23459799 PMCID: PMC3584841 DOI: 10.7150/jca.5836] [Citation(s) in RCA: 55] [Impact Index Per Article: 4.6] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/20/2013] [Accepted: 02/14/2013] [Indexed: 02/06/2023] Open
Abstract
The discovery of microRNA, a group of regulatory short RNA fragments, has added a new dimension to the diagnosis and management of neoplastic diseases. Differential expression of microRNA in a unique pattern in a wide range of tumor types enables researches to develop a microRNA-based assay for source identification of metastatic disease of unknown origin. This is just one example of many microRNA-based cancer diagnostic and prognostic assays in various phases of clinical research. Since colorectal cancer (CRC) is a phenotypic expression of multiple molecular pathways including chromosomal instability (CIN), micro-satellite instability (MIS) and CpG islands promoter hypermethylation (CIMP), there is no one-unique pattern of microRNA expression expected in this disease and indeed, there are multiple reports published, describing different patterns of microRNA expression in CRC. The scope of this manuscript is to provide a comprehensive review of the scientific literature describing the dysregulation of and the potential role for microRNA in the management of CRC. A Pubmed search was conducted using the following MeSH terms, "microRNA" and "colorectal cancer". Of the 493 publications screened, there were 57 papers describing dysregulation of microRNA in CRC.
Collapse
|
18
|
Man YG, Stojadinovic A, Mason J, Avital I, Bilchik A, Bruecher B, Protic M, Nissan A, Izadjoo M, Zhang X, Jewett A. Tumor-infiltrating immune cells promoting tumor invasion and metastasis: existing theories. J Cancer 2013; 4:84-95. [PMID: 23386907 PMCID: PMC3564249 DOI: 10.7150/jca.5482] [Citation(s) in RCA: 134] [Impact Index Per Article: 11.2] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/04/2012] [Accepted: 12/20/2012] [Indexed: 12/12/2022] Open
Abstract
It is a commonly held belief that infiltration of immune cells into tumor tissues and direct physical contact between tumor cells and infiltrated immune cells is associated with physical destructions of the tumor cells, reduction of the tumor burden, and improved clinical prognosis. An increasing number of studies, however, have suggested that aberrant infiltration of immune cells into tumor or normal tissues may promote tumor progression, invasion, and metastasis. Neither the primary reason for these contradictory observations, nor the mechanism for the reported diverse impact of tumor-infiltrating immune cells has been elucidated, making it difficult to judge the clinical implications of infiltration of immune cells within tumor tissues. This mini-review presents several existing hypotheses and models that favor the promoting impact of tumor-infiltrating immune cells on tumor invasion and metastasis, and also analyzes their strength and weakness.
Collapse
|
19
|
Brücher BLDM, Bilchik A, Nissan A, Avital I, Stojadinovic A. Tumor response criteria: are they appropriate? Future Oncol 2012; 8:903-6. [PMID: 22894663 DOI: 10.2217/fon.12.78] [Citation(s) in RCA: 5] [Impact Index Per Article: 0.4] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/21/2022] Open
|
20
|
Steele S, Bilchik A, Eberhardt J, Kalina P, Nissan A, Johnson E, Avital I, Stojadinovic A. Using machine-learned bayesian belief networks to predict perioperative risk of clostridium difficile infection following colon surgery. Interact J Med Res 2012; 1:e6. [PMID: 23611947 PMCID: PMC3626137 DOI: 10.2196/ijmr.2131] [Citation(s) in RCA: 7] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/18/2012] [Revised: 06/11/2012] [Accepted: 06/11/2012] [Indexed: 01/18/2023] Open
Abstract
BACKGROUND Clostridium difficile (C-Diff) infection following colorectal resection is an increasing source of morbidity and mortality. OBJECTIVE We sought to determine if machine-learned Bayesian belief networks (ml-BBNs) could preoperatively provide clinicians with postoperative estimates of C-Diff risk. METHODS We performed a retrospective modeling of the Nationwide Inpatient Sample (NIS) national registry dataset with independent set validation. The NIS registries for 2005 and 2006 were used for initial model training, and the data from 2007 were used for testing and validation. International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) codes were used to identify subjects undergoing colon resection and postoperative C-Diff development. The ml-BBNs were trained using a stepwise process. Receiver operating characteristic (ROC) curve analysis was conducted and area under the curve (AUC), positive predictive value (PPV), and negative predictive value (NPV) were calculated. RESULTS From over 24 million admissions, 170,363 undergoing colon resection met the inclusion criteria. Overall, 1.7% developed postoperative C-Diff. Using the ml-BBN to estimate C-Diff risk, model AUC is 0.75. Using only known a priori features, AUC is 0.74. The model has two configurations: a high sensitivity and a high specificity configuration. Sensitivity, specificity, PPV, and NPV are 81.0%, 50.1%, 2.6%, and 99.4% for high sensitivity and 55.4%, 81.3%, 3.5%, and 99.1% for high specificity. C-Diff has 4 first-degree associates that influence the probability of C-Diff development: weight loss, tumor metastases, inflammation/infections, and disease severity. CONCLUSIONS Machine-learned BBNs can produce robust estimates of postoperative C-Diff infection, allowing clinicians to identify high-risk patients and potentially implement measures to reduce its incidence or morbidity.
Collapse
|
21
|
Stojadinovic A, Bilchik A, Smith D, Eberhardt JS, Ward EB, Nissan A, Johnson EK, Protic M, Peoples GE, Avital I, Steele SR. Clinical decision support and individualized prediction of survival in colon cancer: bayesian belief network model. Ann Surg Oncol 2012; 20:161-74. [PMID: 22899001 DOI: 10.1245/s10434-012-2555-4] [Citation(s) in RCA: 43] [Impact Index Per Article: 3.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/01/2012] [Indexed: 12/16/2022]
Abstract
BACKGROUND We used a large population-based data set to create a clinical decision support system (CDSS) for real-time estimation of overall survival (OS) among colon cancer (CC) patients. Patients with CC diagnosed between 1969 and 2006 were identified from the Surveillance Epidemiology and End Results (SEER) registry. Low- and high-risk cohorts were defined. The tenfold cross-validation assessed predictive utility of the machine-learned Bayesian belief network (ml-BBN) model for clinical decision support (CDS). METHODS A data set consisting of 146,248 records was analyzed using ml-BBN models to provide CDS in estimating OS based on prognostic factors at 12-, 24-, 36-, and 60-month post-treatment follow-up. RESULTS Independent prognostic factors in the ml-BBN model included age, race; primary tumor histology, grade and location; Number of primaries, AJCC T stage, N stage, and M stage. The ml-BBN model accurately estimated OS with area under the receiver-operating-characteristic curve of 0.85, thereby improving significantly upon existing AJCC stage-specific OS estimates. Significant differences in OS were found between low- and high-risk cohorts (odds ratios for mortality: 17.1, 16.3, 13.9, and 8.8 for 12-, 24-, 36-, and 60-month cohorts, respectively). CONCLUSIONS A CDSS was developed to provide individualized estimates of survival in CC. This ml-BBN model provides insights as to how disease-specific factors influence outcome. Time-dependent, individualized mortality risk assessments may inform treatment decisions and facilitate clinical trial design.
Collapse
|
22
|
Stojadinovic A, Nissan A, Wainberg Z, Shen P, McCarter M, Protic M, Howard RS, Steele SR, Peoples GE, Bilchik A. Time-dependent trends in lymph node yield and impact on adjuvant therapy decisions in colon cancer surgery: an international multi-institutional study. Ann Surg Oncol 2012; 19:4178-85. [PMID: 22805869 DOI: 10.1245/s10434-012-2501-5] [Citation(s) in RCA: 11] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/14/2012] [Indexed: 12/12/2022]
Abstract
BACKGROUND Lymph node yield (LNY) and accuracy of nodal assessment are critical to staging and treatment planning in colon cancer (CC). A nationally agreed upon 12-node minimum is a quality standard in CC. The impact of this quality measure on LNY and impact on therapeutic decisions are evaluated in two international, multi-center, prospective trials comprising a well-characterized cohort assembled over 8 years (2001-2009) with long-term follow-up. HYPOTHESIS Quality adherence through increased LNY improves staging accuracy and impacts adjuvant therapy decisions. METHODS Retrospective analysis of prospective data to assess time-dependent LNY, the dependent variable in multivariate linear regression analysis adjusted for age, gender, body-mass-index (BMI), tumor size/stage/grade, anatomic location and surgery date. RESULTS Two-hundred-forty-five patients with non-metastatic CC, median age 70 years, BMI 26 kg/m(2), tumor size 4.0 cm, and LNY 17 nodes were studied. Seventy-two percent had T3 (70 %)/T4 (2 %) tumors. Adherence to the 12-node minimum was 70 %(2001-2002), 81 % (2003-2004), 90 % (2005-2006), 94 % (2007-2008). LNY significantly increased over time (Median LNY: 2001-2004 = 15 vs. 2005-2008 = 17; P < 0.001) on multivariate analysis controlling for tumor size (P < 0.001), and right-sided tumor location (P < 0.001). Adjuvant therapy administration and indication for chemotherapy according to LNY (<12 vs. 12 + LNs = 33 % vs. 39 %; P = 0.48) and time period (2001-2004 vs. 2005-2008 = 39 % vs. 37 %; P = 0.89) remained unchanged. CONCLUSIONS Despite the independent predictors of nodal yield (tumor location and size), year of study still had a significant impact on nodal yield. Despite increased quality adherence and LNY over time, there appears to be a delayed impact on adjuvant therapy decisions once quality standard adherence takes effect.
Collapse
|
23
|
Eberhardt J, Bilchik A, Stojadinovic A. Clinical decision support systems: potential with pitfalls. J Surg Oncol 2012; 105:502-10. [PMID: 22441903 DOI: 10.1002/jso.23053] [Citation(s) in RCA: 20] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/24/2023]
Abstract
Clinical Decision Support Systems (CDSS), an important part of clinical practice, are comprised of a: knowledge base; program for integrating patient-specific information with the knowledge-base; and, user-interface to allow clinicians to interact with the system and get the right information needed to make the right decision for the right patient at the right time. We review the common approaches to CDSS, their strengths and weaknesses and how they are evaluated and developed for clinical use.
Collapse
|
24
|
Short SS, Stojadinovic A, Nissan A, Wainberg Z, Dhall D, Yao K, Bilchik A. Adjuvant treatment of early colon cancer with micrometastases: results of a national survey. J Surg Oncol 2012; 106:119-22. [PMID: 22308106 DOI: 10.1002/jso.23057] [Citation(s) in RCA: 7] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/05/2011] [Accepted: 01/09/2012] [Indexed: 11/10/2022]
Abstract
BACKGROUND AND OBJECTIVES Optimal adjuvant treatment for patients with Stage I/II colon cancer with micrometastases (MM) is unknown. Because there is no known adjuvant treatment-related benefit, we evaluated whether MM influenced treatment decisions. METHOD Review of a national survey from members of the SSO and ASCO. RESULTS Of 602 survey responses, 305 (51%) stated that MM had significant prognostic value, 250 (42%) were unsure, and 47 (7%) did not believe that MM held prognostic value. Three hundred seventy-four (63%) would offer adjuvant therapy in the setting of MM, while 222 (37%) would not. Only 15% routinely performed IHC on lymph nodes. Medical oncologists were more likely to recommend adjuvant therapy compared to surgical oncologists (68% vs. 51%, P = 0.001). CONCLUSIONS MM in colon cancer apparently influenced adjuvant treatment decisions absent known prognostic benefit. Prospective trials are needed to improve the selection of patients for systemic chemotherapy in early, node-negative colon cancer.
Collapse
|
25
|
Bilchik A, Faries MB. Diverting ileostomies: primum minus nocere (first, do less harm). ARCHIVES OF SURGERY (CHICAGO, ILL. : 1960) 2011; 146:1196-1197. [PMID: 22121534 DOI: 10.1001/archsurg.2011.229] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 05/31/2023]
|