Prostate cancer decision aid system and method

Abstract

The present application describes a prostate cancer decision aid system and method. The computer-implemented system determines a treatment for a prostate cancer patient, based on a predictor selection module which is configured to receive a set of patient characteristics, segment the set of patient characteristics in a set of predictors, and output a patient biopsychosocial data set.

Description

PROSTATE CANCER DECISION AID SYSTEM AND METHOD

[0001] The present application describes a prostate cancer decision aid system and method.

[0002] To date, a plethora of prediction models guides management decisions for localized prostate cancer (PCa), visualized in nomograms, lookup tables, categorized point systems, and online calculators [1,2,3].

[0003] There are many treatment options available to patients with localized prostate cancer, including external-beam radiotherapy, brachytherapy, radical prostatectomy, or conservative management strategies for favorable-risk prostate cancer to delay or avoid aggressive treatment and potential side effects. Treatment decisions are complex since they involve uncertainties and trade-offs that must be carefully weighed before choosing. Clinical prediction models can and should be used to guide treatment choices based on the preferred outcomes of individual patients. Despite the large number of prediction models currently available, many more are still needed for many important outcomes, particularly psychological and quality-of-life outcomes, which are critically important in the setting of clinically localized PCa [1]. When considering the impact of disease and treatment on PCa patients’ short- and long-term well-being, the physical consequences, including urinary and sexual dysfunction, are well described (Table 1).

[0004] Table 1. Summary of published prediction models for prognostic risk of sexual or mental health outcomes in patients with localized PCa.ReferenceOutcomesPredictorsBriganti 2010Erectile function at 1 and 3 years (IIEF)Age, baseline IIEF, CCI, BMINovara 2010Erectile function at 12 monthsAge at diagnosis, BMI, CCI, preoperative total PSA, estimated prostate volume, biopsy Gleason score, D'Amico risk scoreJeong 2014Urinary continence at 12 monthsAge at surgery, membranous urethral length, surgery typeHaskins 2014Erectile function (1 question) at 2 after RPAge, BMI, smoker, comorbidities, nerve-sparing / Comorbidities, patient risk, radiotherapy approachMatsushita 2015Urinary continence at 6 and 12 monthsAge, BMI, ASA score, urethral lengthMulhall 2019Erectile function (IIEF) at 3 and 12 months after RPAge; IIEF baseline; nerve-sparing scoring (3-month and 12-month models only); # de comorbidilidades (baseline, developed within the 1st year after surgery)Laviana 2020Sexual, urinary, bowel, and hormonal function (EPIC-26)at 5 years after RP, RT, e ASAge, race, PSA, Gleason score, EPIC baseline (overall health), interactions EPIC * age, treatmentAlemozaffar 2021Erectile function (EPIC) at 2 yearsPretreatment sexual HRQOL score, age, serum PSA level, race / ethnicity, body mass index and intended treatment details.Hasannejadasl 2023Erectile function (EPIC) at 1 and 2 years after treatmentTreatment, baseline quality and frequency erections, tumour stage T e T2, hormonal therapy, comorbidities, fatigue, alcohol consumptionAgochukwu-Mmonu 2023Sexual function (EPIC-26) after RPAge, baseline sexual function / Qualidade da ereção, BMI, diabetes, PSA, Gleason score, prostate volumeIIEF - International Index of Erectile Function; EPIC - Expanded Prostate Index Composite; RP – Radical Prostatectomy; RT – Radiotherapy; CCI - Charlson Comorbidity Index; BMI – Body Mass Index; HRQOL – Health-Related Quality of Life; PSA – Prostate-Specific Antigen; ASA - American Society of Anesthesiologists.

[0005] The present invention describes a computer-implemented system to determine a treatment for a prostate cancer patient, characterized by comprising a predictor selection module, configured to receive a set of patient characteristics, segment the set of patient characteristics in a set of predictors, and output a patient biopsychosocial data set.

[0006] In a proposed embodiment of present invention, the set of patient characteristics comprises at least one of sociodemographic variables and / or life style and medical history and / or sexual situation and / or validated questionaries and / or tumor characteristics.

[0007] Yet in another proposed embodiment of present invention, the sociodemographic variables comprise age, sex, gender, marital status, education, nationality, working status and religious beliefs; and / or the life style and medical history comprise psychiatric problems, mental health status, physical health and drug consumption; and / or the sexual situation comprises sexual orientation, number of sexual partners, sexual behaviors, sexual activity, sexually transmitted diseases, partners sexual health; and / or the validated questionaries comprise International Index of Erectile Function (IIEF), Premature Ejaculation Diagnostic Tool (PEDT), Female Sexual Distress Scale – Revised (FSDS-R), Big Five Inventory (BFI) subscales E and N, Brief Symptom Inventory (BSI-18), Acceptance and Action Questionnaire (AAQ-II), Sexual Dysfunctional Beliefs Questionnaire (QCSD), Sexual Inhibition / Sexual excitation (SIS / SES-SF) (male), Global measures of sexual satisfaction (GMSEX), Global measures of relationship satisfaction (GMREL), World Health Organization Quality-of-Life (WHOQoL), Satisfaction With Life (SWLS); and / or the tumor characteristics comprise Prostate-Specific Antigen (PSA).

[0008] Yet in another proposed embodiment of present invention, the set of predictors comprise at least one predictor for sexual function domain outcome, and / or a predictor for sexual distress domain outcome, and / or a predictor for sexual satisfaction domain outcome, and / or predictor for mental health domain outcome and / or predictor for quality of life domain outcome.

[0009] Yet in another proposed embodiment of present invention, the patient biopsychosocial data set comprises at least one of Sexual function and / or Sexual distress and / or Sexual satisfaction and / or Mental health and / or Quality of life.

[0010] Yet in another proposed embodiment of present invention, the computer-implemented system comprises an outcome prediction model composed by a nomogram and a front-end application.

[0011] Yet in another proposed embodiment of present invention, the computer-implemented system is configured to output values of the event of interest for specific treatment based on the outcome prediction model.

[0012] The present invention also describes a computer-implemented method to determine a treatment for a prostate cancer patient according to the computer-implemented system previously described, characterized by being implemented over a computer network and executed by a remote administrator computer, and comprising the steps of the predictor selection module collecting set of patient characteristics; the predictor selection module segmenting the set of patient characteristics in a set of predictors; the predictor selection module outputting a patient biopsychosocial data set; and outputting an outcome prediction model based on the patient biopsychosocial data set.

[0013] Yet in another proposed embodiment of present invention, the method is implemented over a computer network and executed by a remote administrator computer.

[0014] The present invention also describes a data processing system characterized by comprising the physical means necessary for the execution of the computer-implemented system and computer-implemented method previously described.

[0015] The present invention also describes a computer program, characterized by comprising programming code or instructions suitable for carrying out the computer-implemented system and the computer-implemented method indicated in preceding claims, in which said computer program is stored, and is executed in a data processing system, remote or in-site, for example a server, performing the respective steps previously described.

[0016] The present invention also describes a computer readable physical data storage device, in which the programming code or instructions of the computer program are stored.

[0017] The present invention describes a prostate cancer decision aid system and method.

[0018] The herein disclosed prostate cancer decision aid system and method comprises a series of prediction models from psychosocial outcomes relevant to patients with prostate cancer (i.e., a set combinations of predictor variables that estimate the risk / probability of outcomes) to be used as such or as nomograms and as online risk calculators.

[0019] A nomogram is a graphical calculator that can be provided in paper, and mainly consists of several parallel lines that represent scales for each predictor variable. These are arranged in such a way that, by using straightedge to connect known values of the predictor variable on two lines, an unknown value (a prediction of event of interest) can be read at the point of intersection with another line.

[0020] This model can be used for a set of specific values related to a specific patient, and the model informs about the risk associated with an outcome (e.g., depression, sexual satisfaction).

[0021] For the purpose of the present development, an online risk calculator consists of a front-end web-based interactive tool that results from the application of prediction models. Therefore, a user of such online risk calculator needs to insert a set of known values, one for each of the predictor variables (i.e., characteristics of the patient that are used to predict the outcomes of interest). The information is inserted in a numerical format and / or by choosing several options available (e.g., radio button, dropdown list). The user may then need to click a button to calculate the probability of occurrence of the event of interest.

[0022] Aiding decision tools based on prediction models are ubiquitous in medicine in general from diagnosis to treatment of diseases. Within prostate cancer care, they are used in many stages of the patient’s clinical path. Particularly, they are used to estimate the risk / probability of patients regarding several outcomes, such as survival, recurrence, erectile disfunction, urinary incontinence, among others, given a specific treatment. Therefore, providing a way for patients and their healthcare providers to weigh risks across different treatments, hence guiding the decision-making process considering the patient’s preferences and values. All of the currently available statistical outlets are medically centered approaches to PCa; no psychological variables are included.

[0023] The proposed approach differs from previously known solutions in the sense that considers an additional set of biopsychosocial outcomes, including sexual distress scale (SDS), global measure of sexual satisfaction (GMSEX), global measure of relationship satisfaction (GMREL), mental health brief symptom inventory (BSI), and expands the sexual functioning domains, considering not only erectile function, but also sexual desire or orgasmic function, which are important trade-offs to discuss when treatment is chosen.

[0024] The disclosed technological approach also considers additional risk predictors and confounder variables compared to those considered in the technical background, taking into account not just medical characteristics of the patient and tumour features, but also psychological traits and relationship dimensions as predictors. These include emotional adjustment of patients, sexual beliefs, relationship and sexual life quality, personality traits, or quality of life indexes. This set of predictors are theoretically grounded on psychological models of sexual functioning, representing a holistic approach regarding the recovering of PCa, as compared with medical ones.

[0025] It particularly differs from previous technical approaches since it features novel prediction models, providing patients and healthcare professionals new tools to help discussing treatment options based on biopsychosocial-related quality-of-life outcomes. Most importantly, the present technical disclosure can be used in multidisciplinary teams, including medical doctors as well as mental health professionals who often address PCa recovery.

[0026] Based on the above, the technical advantages of the proposed invention provide guidance in decision making with regard to PCa treatment based on outcomes that are potentially relevant to patients, which are not considered in current decision aids described in the technical background.

[0027] For better understanding of the present application, figures representing preferred embodiments are herein attached which, however, are not intended to limit the technique disclosed herein.Fig.1

[0028] Representation of the development process of the prediction models to estimate probabilities of events of interest.

[0029] With reference to the figures, some embodiments are now described in more detail, which are however not intended to limit the scope of the present application.

[0030] In one of the embodiments of the invention, the prostate cancer decisioning aid system and method comprises the integration of a series of internally validated prediction models for patients with localized PCa that represent a relationship between outcome variables of interest and a combination of predictor variables. Thus, the developed prediction models can be used to predict a future outcome of an individual patient with localized Pca factoring his characteristics (i.e., predictors) and a specific treatment.

[0031] More specifically, the predicted outcomes will be obtained through a combination of the best subset of candidate predictor variables, including age, tumour features, and variables regarding the psychological vulnerability of the patient such as neuroticism, low extraversion and dysfunctional sexual beliefs. The selection of predictors is based on knowledge about important predictors as well as feature selection techniques.

[0032] To this end, the following validated instruments will be used to collect information for the prediction model:

[0033] IIEF: International Index of Erectile Function;

[0034] PEDT: Premature Ejaculation Diagnostic Tool;

[0035] FSDS-R: Female Sexual Distress Scale – Revised, measures the distress associated with impaired sexual function;

[0036] BFI – Big Five Inventory, measures the Big Five Personality Domains;

[0037] BSI: Brief Symptom Inventory, measures psychological symptoms in medical patients;

[0038] AAQII: Acceptance and Action Questionnaire, measures inflexibility / experiential avoidance;

[0039] QCSD: Sexual Dysfunctional Beliefs Questionnaire;

[0040] SIS.SES-SF: Sexual Inhibition / Sexual excitation;

[0041] GMSEX: Global measures of sexual satisfaction;

[0042] GMREL: Global measures of relationship satisfaction;

[0043] SPS: Sensory Processing Sensitivity Questionnaire;

[0044] WHOQoL: World Health Organization Quality-of-Life Questionnaire, measures QoL as an individual’s perception of their position in life in the context of the culture and value systems in which they live and in relation to their goals, and expectations;

[0045] SWLS: Satisfaction With Life, measures global cognitive judgements of satisfaction with one’s life.

[0046] The outcome variables considered for the envisioned prediction models will comprise several domains of the sexual function (IIEF), sexual distress (SDS), sexual satisfaction (GMSEX, GMREL), mental health (BSI) and quality of life (WHOQoL at 6 months after treatment).

[0047] In essence, the ith prediction model will estimate the probability of the ith outcome for a specific patient j (Yij), given the patient’s clinical characteristics at baseline (before treatment) XC1j, …, XCpj, psychosocial characteristics at baseline (before treatment) XP1j, …, XPqj, and treatment T, which can be summarized as follows:

[0048] Yij~ T + XC1j+ … +XCpj+ XP1j+ …+XPqj

[0049] where, the symbol “~” represents the outcome variable as a combination of predictors that provide an estimate of the probability of theithoutcome of interest given specific predictor values.

[0050] In other words, the developed prediction models establish relative effects that each predictor variable has on the outcomes. Each predictor will be converted into a point scale ranging from 0 to 100 and visually represented as nomograms, where the "scores" for each covariate are added together and correspond to the predicted probability of a patient having the outcome of interest.

[0051] The herein proposed system and method for determining a treatment for a prostate cancer patient, in one of the proposed embodiments of the invention, may be implemented using one or more processing units, one or more processing devices, any means for processing, such as a processor, a computer or a programmable hardware component being operable with accordingly adapted software. In other words, the herein described method is then executed on one or more programmable hardware components. Such hardware components may comprise a general-purpose processor, a Digital Signal Processor (DSP), a micro-controller, etc.

[0052] The reference numbers relate to:

[0053] (10) – patients with localized Pca, undergoing treatment Ti;

[0054] (100) – predictor selection;

[0055] (1100) – patient characteristics (candidate predictors);

[0056] (1110) – Sociodemographic variables;

[0057] (1120) – Life style and medical history;

[0058] (1130) – Sexual situation;

[0059] (1140) – Validated questionaries;

[0060] (1150) – Tumor characteristics;

[0061] (1210) – predictors for sexual function domain outcomes;

[0062] (1220) – predictors for sexual distress domain outcomes;

[0063] (1230) – predictors for sexual satisfaction domain outcomes;

[0064] (1240) – predictors for mental health domain outcomes;

[0065] (1250) – predictors for quality of life domain outcomes;

[0066] (1300) – patient biopsychosocial data set (values of predictors);

[0067] (1310) – Sexual function (IIEF);

[0068] (1320) – Sexual distress (SDS);

[0069] (1330) – Sexual satisfaction (GMSEX);

[0070] (1340) – Mental health (BSI);

[0071] (1350) – Quality of life (WHOQoL);

[0072] (200) – prediction model of outcome i;

[0073] (210) – nomogram;

[0074] (220) – front-end application;

[0075] (300) – output values of the event of interest for specific treatment;

[0076] (310) – patient’s risk estimate for treatment 1;

[0077] (320) – patient’s risk estimate for treatment 2;

[0078] (330) – patient’s risk estimate for treatment 3;Non Patent Literature

[0079] 1] Kattan MW. Do we need more nomograms for predicting outcomes in patients with prostate cancer?. Nature Clinical Practice Urology. 2008 Jul;5(7):366-7.

[0080] 2] Shariat SF, Karakiewicz PI, Margulis V, Kattan MW. Inventory of prostate cancer predictive tools. Current opinion in urology. 2008 May 1;18(3):279-96.

[0081] 3] Shariat SF, Karakiewicz PI, Roehrborn CG, Kattan MW. An updated catalog of prostate cancer predictive tools. Cancer: Interdisciplinary International Journal of the American Cancer Society. 2008 Dec 1;113(11):3075-99.

[0082] 4] Briganti A, Gallina A, Suardi N, et al. Predicting erectile function recovery after bilateral nerve sparing radical prostatectomy: a proposal of a novel preoperative risk stratification. The journal of sexual medicine. 2010;7(7):2521-2531.

[0083] 5] Novara G, Ficarra V, D'Elia C, Secco S, De Gobbi A, Cavalleri S et al. Preoperative criteria to select patients for bilateral nerve-sparing robotic-assisted radical prostatectomy. J Sex Med 2010; 7 (2 Pt 1): 839–845.

[0084] 6] Jeong SJ, Yeon JS, Lee JK, Cha WH, Jeong JW, Lee BK et al. Development and validation of nomograms to predict the recovery of urinary continence after radical prostatectomy: comparisons between immediate, early, and late continence. World J Urol 2014; 32: 437–444

[0085] 7] Haskins AE, Han PK, Lucas FL, Bristol I, Hansen M. Development of clinical models for predicting erectile function after localized prostate cancer treatment. International Journal of Urology. 2014;21(12):1227-1233.

[0086] 8] Matsushita K, Kent MT, Vickers AJ, von Bodman C, Bernstein M, Touijer KA et al. Preoperative predictive model of recovery of urinary continence after radical prostatectomy. BJU Int 2015; 116: 577–583.

[0087] 9] Mulhall JP, Kattan MW, Bennett NE, Stasi J, Nascimento B, Eastham J, Guillonneau B, Scardino PT. Development of nomograms to predict the recovery of erectile function following radical prostatectomy. The journal of sexual medicine. 2019 Nov 1;16(11):1796-802.

[0088] 10] Laviana AA, Zhao Z, Huang L-C, et al. Development and internal validation of a web-based tool to predict sexual, urinary, and bowel function longitudinally after radiation therapy, surgery, or observation. European urology. 2020;78(2):248-255.11.

[0089] 11] Alemozaffar M, Regan MM, Cooperberg MR, et al. Prediction of erectile function following treatment for prostate cancer. Jama. 2011;306(11):1205-1214.

[0090] 12] Hasannejadasl H, Roumen C, van der Poel H, et al. Development and external validation of multivariate prediction models for erectile dysfunction in men with localized prostate cancer. Plos one. 2023;18(3):e0276815.

[0091] 13] Agochukwu-Mmonu N, Murali A, Wittmann D, et al. Development and Validation of Dynamic Multivariate Prediction Models of Sexual Function Recovery in Patients with Prostate Cancer Undergoing Radical Prostatectomy: Results from the MUSIC Statewide Collaborative. Eur Urol Open Sci. Jun 2022;40:1-8. doi:10.1016 / j.euros.2022.03.009

Claims

Computer-implemented system to determine a treatment for a prostate cancer patient (10), characterized by comprising a predictor selection module (100), configured to receive a set of patient characteristics (1100), segment the set of patient characteristics (1100) in a set of predictors, and output a patient biopsychosocial data set (1300).Computer-implemented system according to previous claim 1, wherein the set of patient characteristics (1100) comprises at least one of sociodemographic variables (1110) and / or life style and medical history (1120) and / or sexual situation (1130) and / or validated questionaries (1140) and / or tumor characteristics (1150).Computer-implemented system according to previous claim 2, whereinthe sociodemographic variables (1110) comprise age, sex, gender, marital status, education, nationality, working status and religious beliefs; and / orthe life style and medical history (1120) comprise psychiatric problems, mental health status, physical health and drug consumption; and / orthe sexual situation (1130) comprises sexual orientation, number of sexual partners, sexual behaviors, sexual activity, sexually transmitted diseases, partners sexual health; and / orthe validated questionaries (1140) comprise International Index of Erectile Function (IIEF), Premature Ejaculation Diagnostic Tool (PEDT), Female Sexual Distress Scale – Revised (FSDS-R), Big Five Inventory (BFI) subscales E and N, Brief Symptom Inventory (BSI-18), Acceptance and Action Questionnaire (AAQ-II), Sexual Dysfunctional Beliefs Questionnaire (QCSD), Sexual Inhibition / Sexual excitation (SIS / SES-SF) (male), Global measures of sexual satisfaction (GMSEX), Global measures of relationship satisfaction (GMREL), World Health Organization Quality-of-Life (WHOQoL), Satisfaction With Life (SWLS); and / orthe tumor characteristics (1150) comprise Prostate-Specific Antigen (PSA).Computer-implemented system according to previous claim 1, wherein the set of predictors comprise at least one predictor for sexual function domain outcome (1210), and / or a predictor for sexual distress domain outcome (1220), and / or a predictor for sexual satisfaction domain outcome (1230), and / or predictor for mental health domain outcome (1240) and / or predictor for quality of life domain outcome (1250).Computer-implemented system according to previous claim 1, wherein the patient biopsychosocial data set (1300) comprises at least one of Sexual function (1310) and / or Sexual distress (1320) and / or Sexual satisfaction (1330) and / or Mental health (1340) and / or Quality of life (1350).Computer-implemented system according to previous claim 1, comprising an outcome prediction model (200) composed by a nomogram (210) and a front-end application (220).Computer-implemented system according to previous claim 1, configured to output values of the event of interest for specific treatment (1300) based on the outcome prediction model (200).Computer-implemented method to determine a treatment for a prostate cancer patient (10) according to the computer-implemented system described in claim 1 characterized by being implemented over a computer network and executed by a remote administrator computer, and comprising the steps ofthe predictor selection module (100) collecting set of patient characteristics (1100);the predictor selection module (100) segmenting the set of patient characteristics (1100) in a set of predictors;the predictor selection module (100) outputting a patient biopsychosocial data set (1300); andoutputting an outcome prediction model (200) based on the patient biopsychosocial data set (1300).Computer-implemented method for interacting with the computer-implemented system of preceding claims 1 to 7, said method being implemented over a computer network and executed by a remote administrator computer.Data processing system, characterized by comprising the physical means necessary for the execution of the computer- implemented system and computer-implemented method described in any one of the preceding claims.Computer program, characterized by comprising programming code or instructions suitable for carrying out the computer-implemented system and the computer-implemented method indicated in preceding claims, in which said computer program is stored, and is executed in a data processing system, remote or in-site, for example a server, performing the respective steps described in the claims.Computer readable physical data storage device, in which the programming code or instructions of the computer program are stored.

Images