Prediction of Early Distant Recurrence in Upfront Resectable Pancreatic Adenocarcinoma: A Multidisciplinary, Machine Learning-Based Approach

Diego Palumbo, Martina Mori, Francesco Prato, Stefano Crippa, Giulio Belfiori, Michele Reni, Junaid Mushtaq, Francesca Aleotti, Giorgia Guazzarotti, Roberta Cao, Stephanie Steidler, Domenico Tamburrino, Emiliano Spezi, Antonella Del Vecchio, Stefano Cascinu, Massimo Falconi, Claudio Fiorino, Francesco De Cobelli, Diego Palumbo, Martina Mori, Francesco Prato, Stefano Crippa, Giulio Belfiori, Michele Reni, Junaid Mushtaq, Francesca Aleotti, Giorgia Guazzarotti, Roberta Cao, Stephanie Steidler, Domenico Tamburrino, Emiliano Spezi, Antonella Del Vecchio, Stefano Cascinu, Massimo Falconi, Claudio Fiorino, Francesco De Cobelli

Abstract

Despite careful selection, the recurrence rate after upfront surgery for pancreatic adenocarcinoma can be very high. We aimed to construct and validate a model for the prediction of early distant recurrence (<12 months from index surgery) after upfront pancreaticoduodenectomy. After exclusions, 147 patients were retrospectively enrolled. Preoperative clinical and radiological (CT-based) data were systematically evaluated; moreover, 182 radiomics features (RFs) were extracted. Most significant RFs were selected using minimum redundancy, robustness against delineation uncertainty and an original machine learning bootstrap-based method. Patients were split into training (n = 94) and validation cohort (n = 53). Multivariable Cox regression analysis was first applied on the training cohort; the resulting prognostic index was then tested in the validation cohort. Clinical (serum level of CA19.9), radiological (necrosis), and radiomic (SurfAreaToVolumeRatio) features were significantly associated with the early resurge of distant recurrence. The model combining these three variables performed well in the training cohort (p = 0.0015, HR = 3.58, 95%CI = 1.98-6.71) and was then confirmed in the validation cohort (p = 0.0178, HR = 5.06, 95%CI = 1.75-14.58). The comparison of survival curves between low and high-risk patients showed a p-value <0.0001. Our model may help to better define resectability status, thus providing an actual aid for pancreatic adenocarcinoma patients' management (upfront surgery vs. neoadjuvant chemotherapy). Independent validations are warranted.

Keywords: X-ray; computed tomography; machine learning; pancreatic adenocarcinoma; prognosis; radiomics.

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1
Figure 1
Inclusion and exclusion criteria flowchart.
Figure 2
Figure 2
Radiomic features extraction workflow.
Figure 3
Figure 3
10 most frequent variables identified through a machine-learning bootstrap-ranking procedure (those retaining p value less than 0.05 in more than 500 cases on the 1000 bootstrapped samples): eight radiomic features (3 morphologic, 4 texture related and 1 statistical features) and two clinicoradiological variables (radiological necrosis, serum level of CA19.9).
Figure 4
Figure 4
Radiomic model overall performance in terms of (a) area under the ROC curve (AUC) for both training (red empty circles) and validation (red filled squares) cohorts, and outcome prediction in terms of Kaplan Meier curve separation between low and high risk patients according to the computed prognostic index in both training (b) and validation (c) cohorts.
Figure 5
Figure 5
Clinicoradiological model overall performance in terms of (a) area under the ROC curve (AUC) for both training (green empty circles) and validation (green filled squares) cohorts, and outcome prediction in terms of Kaplan Meier curve separation between low and high risk patients according to the computed prognostic index in both training (b) and validation (c) cohorts.
Figure 6
Figure 6
Combined model overall performance in terms of (a) area under the ROC curve (AUC) for both training (blue empty circles) and validation (blue filled squares) cohorts, and outcome prediction in terms of Kaplan Meier curve separation between low and high risk patients according to the computed prognostic index in both training (b) and validation (c) cohorts.

References

    1. Zins M., Matos C., Cassinotto C. Pancreatic Adenocarcinoma Staging in the Era of Preoperative Chemotherapy and Radiation Therapy. Radiology. 2018;287:374–390. doi: 10.1148/radiol.2018171670.
    1. National Comprehensive Cancer Network . NCCN Clinical Practice Guidelines in Oncology. Pancreatic Adenocarcinoma. National Comprehensive Cancer Network; Fort Washington, PA, USA: 2019. Version 1.2019.
    1. Petrelli F., Inno A., Barni S., Ghidini A., Labianca R., Falconi M., Reni M., Cascinu S. Borderline resectable pancreatic cancer: More than an anatomical concept. Dig. Liver Dis. 2017;49:223–226. doi: 10.1016/j.dld.2016.11.010.
    1. Matsumoto I., Murakami Y., Shinzeki M., Shinzeki M., Asari S., Goto T., Tani M., Motoi F., Uemura K., Sho M., et al. Proposed preoperative risk factors for early recurrence in patients with resectable pancreatic ductal adenocarcinoma after surgical resection: A multi-center retrospective study. Pancreatology. 2015;15:674–680. doi: 10.1016/j.pan.2015.09.008.
    1. Lee Y.S., Lee J.C., Yang S.Y., Kim J., Hwang J.H. Neoadjuvant therapy versus upfront surgery in resectable pancreatic cancer according to intention-to-treat and per-protocol analysis: A systematic review and meta-analysis. Sci. Rep. 2019;9:15662. doi: 10.1038/s41598-019-52167-9.
    1. Barugola G., Partelli S., Marcucci S., Sartori N., Capelli P., Bassi C., Pederzoli P., Falconi M. Resectable pancreatic cancer: Who really benefits from resection? Ann. Surg. Oncol. 2009;16:3316–3322. doi: 10.1245/s10434-009-0670-7.
    1. Bergquist J.R., Puig C.A., Shubert C.R., Groeschl R.T., Habermann E.B., Kendrick M.L., Nagorney D.M., Smoot R.L., Farnell M.B., Truty M.J. Carbohydrate Antigen 19-9 Elevation in Anatomically Resectable, Early Stage Pancreatic Cancer Is Independently Associated with Decreased Overall Survival and an Indication for Neoadjuvant Therapy: A National Cancer Database Study. JACS. 2016;223:52–65. doi: 10.1016/j.jamcollsurg.2016.02.009.
    1. Nakamura T., Asano T., Okamura K., Tsuchikawa T., Murakami S., Kurashima Y., Ebihara Y., Noji T., Nakanishi Y., Tanaka K.A. A Preoperative Prognostic Scoring System to Predict Prognosis for Resectable Pancreatic Cancer: Who Will Benefit from Upfront Surgery? J. Gastrointest. Surg. 2019;23:990–996. doi: 10.1007/s11605-018-3972-x.
    1. Groot V.P., Rezaee N., Wu W., Cameron J.L., Fishman E.K., Hruban R.H., Weiss M.J., Zheng L., Wolfgang C.L., He J. Patterns, Timing, and Predictors of Recurrence Following Pancreatectomy for Pancreatic Ductal Adenocarcinoma. Ann. Surg. 2018;267:936–945. doi: 10.1097/SLA.0000000000002234.
    1. Barugola G., Frulloni L., Salvia R., Falconi M. Is CA 19-9 a screening marker? Dig. Liver Dis. 2009;41:325–327. doi: 10.1016/j.dld.2009.02.009.
    1. Ballehaninna U.K., Chamberlain R.S. The clinical utility of serum CA 19-9 in the diagnosis, prognosis and management of pancreatic adenocarcinoma: An evidence based appraisal. J. Gastrointest. Oncol. 2012;3:105–119. doi: 10.3978/j.issn.2078-6891.2011.021.
    1. Crippa S., Pergolini I., Javed A.A., Honselmann K.C., Weiss M.J., Di Salvo F., Burkhart R., Zamboni G., Belfiori G., Ferrone C.R. Implications of Perineural Invasion on Disease Recurrence and Survival After Pancreatectomy for Pancreatic Head Ductal Adenocarcinoma. Ann. Surg. 2020 doi: 10.1097/SLA.0000000000004464.
    1. Lambin P., Leijenaar R.T.H., Deist T.M., Peerlings J., de Jong E., van Timmeren J., Sanduleanu S., Larue R., Even A., Jochems A. Radiomics: The bridge between medical imaging and personalized medicine. Nat. Rev. Clin. Oncol. 2017;14:749–762. doi: 10.1038/nrclinonc.2017.141.
    1. Abunahel B.M., Pontre B., Kumar H., Petrov M.S. Pancreas image mining: A systematic review of radiomics. Eur. Radiol. 2021;31:3447–3467. doi: 10.1007/s00330-020-07376-6.
    1. Zwanenburg A., Vallières M., Abdalah M.A., Aerts H., Andrearczyk V., Apte A., Ashrafinia S., Bakas S., Beukinga R.J., Boellaard R. The Image Biomarker Standardization Initiative: Standardized Quantitative Radiomics for High-Throughput Image-based Phenotyping. Radiology. 2020;295:328–338. doi: 10.1148/radiol.2020191145.
    1. Loi S., Mori M., Benedetti G., Partelli S., Broggi S., Cattaneo G.M., Palumbo D., Muffatti F., Falconi M., De Cobelli F. Robustness of CT radiomic features against image discretization and interpolation in characterizing pancreatic neuroendocrine neoplasms. Phys. Med. 2020;76:125–133. doi: 10.1016/j.ejmp.2020.06.025.
    1. D’Onofrio M., De Robertis R., Aluffi G., Cadore C., Beleù A., Cardobi N., Malleo G., Manfrin E., Bassi C. CT Simplified Radiomic Approach to Assess the Metastatic Ductal Adenocarcinoma of the Pancreas. Cancers (Basel) 2021;13:1843. doi: 10.3390/cancers13081843.
    1. Mori M., Passoni P., Incerti E., Bettinardi V., Broggi S., Reni M., Whybra P., Spezi E., Vanoli E.G., Gianolli L. Training and validation of a robust PET radiomic-based index to predict distant-relapse-free-survival after radio-chemotherapy for locally advanced pancreatic cancer. Radiother. Oncol. 2020;153:258–264. doi: 10.1016/j.radonc.2020.07.003.
    1. Van Timmeren J.E., Cester D., Tanadini-Lang S., Alkadhi H., Baessler B. Radiomics in medical imaging-"how-to" guide and critical reflection. Insights Imaging. 2020;11:91. doi: 10.1186/s13244-020-00887-2.
    1. Vallières M., Zwanenburg A., Badic B., Cheze Le Rest C., Visvikis D., Hatt M. Responsible Radiomics Research for Faster Clinical Translation. J. Nucl. Med. 2018;59:189–193. doi: 10.2967/jnumed.117.200501.
    1. Choi S.H., Kim H.J., Kim K.W., An S., Hong S.M., Kim S.C., Kim M.H. DPC4 gene expression in primary pancreatic ductal adenocarcinoma: Relationship with CT characteristics. Br. J. Radiol. 2017;90:20160403. doi: 10.1259/bjr.20160403.
    1. Collins G.S., Reitsma J.B., Altman D.G., Moons K.G. Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD): The TRIPOD statement. Ann. Intern Med. 2015;162:55–63. doi: 10.7326/M14-0697.
    1. Groot V.P., Gemenetzis G., Blair A.B., Rivero-Soto R.J., Yu J., Javed A.A., Burkhart R.A., Rinkes I., Molenaar I.Q., Cameron J.L. Defining and Predicting Early Recurrence in 957 Patients With Resected Pancreatic Ductal Adenocarcinoma. Ann. Surg. 2019;269:1154–1162. doi: 10.1097/SLA.0000000000002734.
    1. Yamamoto Y., Ikoma H., Morimura R., Konishi H., Murayama Y., Komatsu S., Shiozaki A., Kuriu Y., Kubota T., Nakanishi M. Optimal duration of the early and late recurrence of pancreatic cancer after pancreatectomy based on the difference in the prognosis. Pancreatology. 2014;14:524–529. doi: 10.1016/j.pan.2014.09.006.
    1. Braga M., Pecorelli N., Ariotti R., Capretti G., Greco M., Balzano G., Castoldi R., Beretta L. Enhanced recovery after surgery pathway in patients undergoing pancreaticoduodenectomy. World J. Surg. 2014;38:2960–2966. doi: 10.1007/s00268-014-2653-5.
    1. Crippa S., Belfiori G., Tamburrino D., Partelli S., Falconi M. Indications to total pancreatectomy for positive neck margin after partial pancreatectomy: A review of a slippery ground. Updates Surg. 2021;73:1219–1229. doi: 10.1007/s13304-021-01141-0.
    1. Van Roessel S., Kasumova G.G., Verheij J., Najarian R.M., Maggino L., de Pastena M., Malleo G., Marchegiani G., Salvia R., Ng S.C. International Validation of the Eighth Edition of the American Joint Committee on Cancer (AJCC) TNM Staging System in Patients with Resected Pancreatic Cancer. JAMA Surg. 2018;153:e183617. doi: 10.1001/jamasurg.2018.3617.
    1. Palumbo D., Tamburrino D., Partelli S., Gusmini S., Guazzarotti G., Cao R., Crippa S., Falconi M., De Cobelli F. Before sentinel bleeding: Early prediction of postpancreatectomy hemorrhage (PPH) with a CT-based scoring system. Eur. Radiol. 2021;31:6879–6888. doi: 10.1007/s00330-021-07788-y.
    1. Al-Hawary M.M., Francis I.R., Chari S.T., Fishman E.K., Hough D.M., Lu D.S., Macari M., Megibow A.J., Miller F.H., Mortele K.J. Pancreatic ductal adenocarcinoma radiology reporting template: Consensus statement of the society of abdominal radiology and the american pancreatic association. Gastroenterology. 2014;146:291–304.e1. doi: 10.1053/j.gastro.2013.11.004.
    1. Benedetti G., Mori M., Panzeri M.M., Barbera M., Palumbo D., Sini C., Muffatti F., Andreasi V., Steidler S., Doglioni C., et al. CT-derived radiomic features to discriminate histologic characteristics of pancreatic neuroendocrine tumors. Radiol. Med. 2021;126:745–760. doi: 10.1007/s11547-021-01333-z.
    1. Peduzzi P., Concato J., Kemper E., Holford T.R., Feinstein A.R. A simulation study of the number of events per variable in logistic regression analysis. J. Clin. Epidemiol. 1996;49:1373–1379. doi: 10.1016/S0895-4356(96)00236-3.
    1. Mori M., Benedetti G., Partelli S., Sini C., Andreasi V., Broggi S., Barbera M., Cattaneo G.M., Muffatti F., Panzeri M. Ct radiomic features of pancreatic neuroendocrine neoplasms (panNEN) are robust against delineation uncertainty. Phys. Med. 2019;57:41–46. doi: 10.1016/j.ejmp.2018.12.005.
    1. Reni M., Balzano G., Zanon S., Zerbi A., Rimassa L., Castoldi R., Pinelli D., Mosconi S., Doglioni C., Chiaravalli M. Safety and efficacy of preoperative or postoperative chemotherapy for resectable pancreatic adenocarcinoma (PACT-15): A randomised, open-label, phase 2-3 trial. Lancet Gastroenterol. Hepatol. 2018;3:413–423. doi: 10.1016/S2468-1253(18)30081-5.
    1. Paniccia A., Hosokawa P., Henderson W., Schulick R.D., Edil B.H., McCarter M.D., Gajdos C. Characteristics of 10-Year Survivors of Pancreatic Ductal Adenocarcinoma. JAMA Surg. 2015;150:701–710. doi: 10.1001/jamasurg.2015.0668.
    1. Kudo M., Kobayashi T., Gotohda N., Konishi M., Takahashi S., Kobayashi S., Sugimoto M., Okubo S., Martin J., Cabral H. Clinical Utility of Histological and Radiological Evaluations of Tumor Necrosis for Predicting Prognosis in Pancreatic Cancer. Pancreas. 2020;49:634–641. doi: 10.1097/MPA.0000000000001539.
    1. Brennan M.F., Kattan M.W., Klimstra D., Conlon K. Prognostic nomogram for patients undergoing resection for adenocarcinoma of the pancreas. Ann. Surg. 2004;240:293–298. doi: 10.1097/01.sla.0000133125.85489.07.
    1. Adamu M., Nitschke P., Petrov P., Rentsch A., Distler M., Reissfelder C., Welsch T., Saeger H.D., Weitz J., Rahbari N.N. Validation of prognostic risk scores for patients undergoing resection for pancreatic cancer. Pancreatology. 2018;18:585–591. doi: 10.1016/j.pan.2018.05.005.
    1. Strijker M., Chen J.W., Mungroop T.H., Jamieson N.B., van Eijck C.H., Steyerberg E.W., Wilmink J.W., Groot Koerkamp B., van Laarhoven H.W., Besselink M.G. Systematic review of clinical prediction models for survival after surgery for resectable pancreatic cancer. Br. J. Surg. 2019;106:342–354. doi: 10.1002/bjs.11111.
    1. Kim D.W., Lee S.S., Kim S.O., Kim J.H., Kim H.J., Byun J.H., Yoo C., Kim K.P., Song K.B., Kim S.C. Estimating Recurrence after Upfront Surgery in Patients with Resectable Pancreatic Ductal Adenocarcinoma by Using Pancreatic CT: Development and Validation of a Risk Score. Radiology. 2020;296:541–551. doi: 10.1148/radiol.2020200281.
    1. Eilaghi A., Baig S., Zhang Y., Zhang J., Karanicolas P., Gallinger S., Khalvati F., Haider M.A. CT texture features are associated with overall survival in pancreatic ductal adenocarcinoma—A quantitative analysis. BMC Med. Imaging. 2017;17:38. doi: 10.1186/s12880-017-0209-5.
    1. Kim Y.C., Kim H.J., Park J.H., Park D.I., Cho Y.K., Sohn C.I., Jeon W.K., Kim B.I., Shin J.H. Can preoperative CA19-9 and CEA levels predict the resectability of patients with pancreatic adenocarcinoma? J. Gastroenterol. Hepatol. 2009;24:1869–1875. doi: 10.1111/j.1440-1746.2009.05935.x.
    1. Ferrone C.R., Finkelstein D.M., Thayer S.P., Muzikansky A., Fernandez del Castillo C., Warshaw A.L. Perioperative CA19-9 levels can predict stage and survival in patients with resectable pancreatic adenocarcinoma. J. Clin. Oncol. 2006;24:2897–2902. doi: 10.1200/JCO.2005.05.3934.
    1. Sugiura T., Uesaka K., Kanemoto H., Mizuno T., Sasaki K., Furukawa H., Matsunaga K., Maeda A. Serum CA19-9 is a significant predictor among preoperative parameters for early recurrence after resection of pancreatic adenocarcinoma. J. Gastrointest. Surg. 2012;16:977–985. doi: 10.1007/s11605-012-1859-9.
    1. Limkin E.J., Reuzé S., Carré A., Sun R., Schernberg A., Alexis A., Deutsch E., Ferté C., Robert C. The complexity of tumor shape, spiculatedness, correlates with tumor radiomic shape features. Sci. Rep. 2019;9:4329. doi: 10.1038/s41598-019-40437-5.
    1. Bribiesca E. Measure of compactness for 3D shapes. Comput. Math. Appl. 2000;40:1275–1284. doi: 10.1016/S0898-1221(00)00238-8.
    1. Harris L.K., Theriot J.A. Surface Area to Volume Ratio: A Natural Variable for Bacterial Morphogenesis. Trends Microbiol. 2018;26:815–832. doi: 10.1016/j.tim.2018.04.008.
    1. Xie T., Wang X., Li M., Tong T., Yu X., Zhou Z. Pancreatic ductal adenocarcinoma: A radiomics nomogram outperforms clinical model and TNM staging for survival estimation after curative resection. Eur. Radiol. 2020;30:2513–2524. doi: 10.1007/s00330-019-06600-2.

Source: PubMed

赞助者和合作者

医疗条件

药物干预

3
订阅