Predicting future cancer burden in the United States by artificial neural networks

Towards reliable forecasting of healthcare capacity needs: A scoping review and evidence mapping

BACKGROUND

The COVID-19 pandemic has highlighted the critical importance of robust healthcare capacity planning and preparedness for emerging crises. However, healthcare systems must also adapt to more gradual temporal changes in disease prevalence and demographic composition over time. To support proactive healthcare planning, statistical capacity forecasting models can provide valuable information to healthcare planners. This systematic literature review and evidence mapping aims to identify and describe studies that have used statistical forecasting models to estimate healthcare capacity needs within hospital settings.

METHOD

Studies were identified in the databases MEDLINE and Embase and screened for relevance before items were defined and extracted within the following categories: forecast methodology, measure of capacity, forecast horizon, healthcare setting, target diagnosis, validation methods, and implementation.

RESULTS

84 studies were selected, all focusing on various capacity outcomes, including number of hospital beds/ patients, staffing, and length of stay. The selected studies employed different analytical models grouped in six items; discrete event simulation (N = 13, 15 %), generalized linear models (N = 21, 25 %), rate multiplication (N = 15, 18 %), compartmental models (N = 14, 17 %), time series analysis (N = 22, 26 %), and machine learning not otherwise categorizable (N = 12, 14 %). The review further provides insights into disease areas with infectious diseases (N = 24, 29 %) and cancer (N = 12, 14 %) being predominant, though several studies forecasted healthcare capacity needs in general (N = 24, 29 %). Only about half of the models were validated using either temporal validation (N = 39, 46 %), cross-validation (N = 2, 2 %) or/and geographical validation (N = 4, 5 %).

CONCLUSION

The forecasting models' applicability can serve as a resource for healthcare stakeholders involved in designing future healthcare capacity estimation. The lack of routine performance validation of the used algorithms is concerning. There is very little information on implementation and follow-up validation of capacity planning models.

Apalutamide or enzalutamide in castration-sensitive prostate cancer: a number needed to treat analysis

The treatment of castration-sensitive prostate cancer (CSPC) has been revolutionized by the advent of apalutamide and enzalutamide in this setting; however, a direct comparison between these agents is still missing. In the current paper, we performed both Number Needed to Treat (NNT) and Number Needed to Harm (NNH) analyses aimed to compare clinical outcomes in CSPC patients treated with apalutamide or enzalutamide; data from 3323 CSPC patients enrolled in the ARCHES, ENZAMET and TITAN phase III studies were included. According to our results, apalutamide showed better results in terms of overall survival (OS) and safety in patients with CSPC, while better outcomes were observed with enzalutamide in the low-volume subgroup.

A Predictive Model for Qualitative Evaluation of PG-SGA in Tumor Patients Through Machine Learning

Objective

Patient-Generated Subjective Global Assessment (PG-SGA) was a nutritional status assessment technique specifically tailored for patients with oncology. The goal of this study was to develop a machine learning (ML) prediction model for predicting PG-SGA categorization of patients with tumor.

Methods

From 2014 to 2020, patients at the First Hospital of Jilin University performed laboratory testing, bioelectrical impedance, physical measures, and the PG-SGA scale. A total of 8230 patients were involved in the study. Patients with missing or partial data were removed, leaving 7287 patients, of which 3743 were males and 3544 were females. ML was used to design a clinical prediction model for PG-SGA categories.

Results

Through the least absolute shrinkage and selection operator (LASSO) and the correlation matrix, 135 variables were screened and 6 variables were retained; ML was performed among the remaining variables. The accuracy of neural network prediction models was 70.3% and 70.4% for males and females in the training cohort, respectively, and 74.4% and 73.2% for males and females in the validation cohort, respectively. The area under curve (AUC) of males was 0.87 for PG-SGA scores “0–3”, 0.70 for PG-SGA scores “4–8” and 0.74 for PG-SGA scores “>8”. As for females, the AUC was 0.85 for PG-SGA scores “0–3”, 0.65 for PG-SGA scores “4–8” and 0.76 for PG-SGA scores “>8”. The results of confusion matrix showed that the models were of good predictive validity. The prediction model was nearly 90% accurate for predictions that do not require nutritional support.

Conclusion

We demonstrated that neural network learning is the best clinical prediction model using ML. The model can work as a prediction for the PG-SGA classification of patients with cancer and can be promoted further in the clinic.

Cost-Effectiveness Analysis of Abemaciclib Plus Fulvestrant in the Second-Line Treatment of Women With HR+/HER2- Advanced or Metastatic Breast Cancer: A US Payer Perspective

Introduction: This study evaluated the cost-effectiveness of abemaciclib plus fulvestrant (ABE + FUL) vs. palbociclib plus fulvestrant (PAL + FUL), ribociclib plus fulvestrant (RIB + FUL) and fulvestrant monotherapy (FUL) as second-line treatment for hormone receptor-positive and human epidermal growth factor receptor 2- negative advanced or metastatic breast cancer in the US.

Methods: The 3 health states partitioned survival (PS) model was used over the lifetime. Effectiveness and safety data were derived from the MONARCH 2 trial, MONALEESA-3 trial, and PALOMA-3 trial. Parametric survival models were used for four treatments to explore the long-term effect. Costs were derived from the pricing files of Medicare and Medicaid Services, and utility values were derived from published studies. Sensitivity analyses including one-way sensitivity analysis, probabilistic sensitivity analysis and scenario analysis were performed to observe model stability.

Results: In the PS model, compared with PAL + FUL, ABE + FUL yielded 0.44 additional QALYs at an additional cost of $100,696 for an incremental cost-utility ratio (ICUR) of $229,039/QALY. Compared with RIB + FUL, ABE + FUL yielded 0.03 additional QALYs at an additional cost of $518 for an ICUR of $19,314/QALY. Compared with FUL, ABE + FUL yielded 0.68 additional QALYs at an additional cost of $260,584 for ICUR of $381,450/QALY. From the PS model, the ICUR was $270,576 /QALY (ABE + FUL vs. PAL + FUL), dominated (ABE + FUL vs. RIB + FUL) and $404,493/QALY (ABE + FUL vs. FUL) in scenario analysis. In the probabilistic sensitivity analysis, the probabilities that ABE + FUL was cost-effective vs. PAL + FUL, RIB + FUL and FUL at thresholds of $50,000, $100,000, and $200,000 per QALY gained were 0% and the probabilities that ABE + FUL was cost-effective vs. PAL + FUL and RIB + FUL at thresholds of $50,000, $100,000, and $200,000 per QALY gained were 0.2, 0.6, and 7.3%.

Conclusions: The findings from the present analysis suggest that ABE + FUL might be cost-effective compared with RIB + FUL and not cost-effective compared with PAL + FUL and FUL for second-line treatment of patients with HR+/HER2– advanced or metastatic breast cancer in the US.

Narrative review: predicting future molecular and clinical profiles of prostate cancer in the United States

Prostate cancer represents the most frequent tumor in men, accounting for the 21% of all diagnosed tumors, with 191,930 new cases and 33,330 deaths estimated in 2020. Advanced prostate cancer represents a heterogeneous disease, ranging from hormone naive or hormone sensitive to castration resistant. The therapeutic armamentarium for this disease has been implemented in the last years by novel hormonal therapies and chemotherapies. However, the percentage of patients who achieve complete responses still results negligible. On this scenario, the design of clinical trials investigating new therapeutic approaches represent a dramatic medical need. Predicting cancer incidence may be fundamental to design specific clinical trials, to optimize the allocation of economic resources, and to plan future cancer control programs. ERG, SPOP and DDR genes alterations can act as therapeutic targets in prostate cancer patients and can be tested to identify a gene-selected patient population to enrol in specific trials. According to our predictions, ERG gene fusions will be the most predominant molecular subtype, accounting for 69,050 new cases in 2030. Mutation in SPOP gene will be diagnosed in 16,512 tumors, corresponding to the number of cases associated with alterations in DDR genes (including 7,956 BRCA2 mutated tumors). In this article, we analyzed and discussed the future molecular and clinical profiles of prostate cancer in the United States, aimed to describe a series of distinct subpopulations and to quantify potential clinical trial candidates in the next years.

Treating Prostate Cancer by Antibody-Drug Conjugates

Prostate cancer is the most frequent malignancy in the worldwide male population; it is also one of the most common among all the leading cancer-related death causes. In the last two decades, the therapeutic scenario of metastatic castration-resistant prostate cancer has been enriched by the use of chemotherapy and androgen receptor signaling inhibitors (ARSI) and, more recently, by immunotherapy and poly(ADP–ribose) polymerase (PARP) inhibitors. At the same time, several trials have shown the survival benefits related to the administration of novel ARSIs among patients with non-castration-resistant metastatic disease along with nonmetastatic castration-resistant cancer too. Consequently, the therapeutic course of this malignancy has been radically expanded, ensuring survival benefits never seen before. Among the more recently emerging agents, the so-called “antibody–drug conjugates” (ADCs) are noteworthy because of their clinical practice changing outcomes obtained in the management of other malignancies (including breast cancer). The ADCs are novel compounds consisting of cytotoxic agents (also known as the payload) linked to specific antibodies able to recognize antigens expressed over cancer cells’ surfaces. As for prostate cancer, researchers are focusing on STEAP1, TROP2, PSMA, CD46 and B7-H3 as optimal antigens which may be targeted by ADCs. In this paper, we review the pivotal trials that have currently changed the therapeutic approach to prostate cancer, both in the nonmetastatic castration-resistant and metastatic settings. Therefore, we focus on recently published and ongoing trials designed to investigate the clinical activity of ADCs against prostate malignancy, characterizing these agents. Lastly, we briefly discuss some ADCs-related issues with corresponding strategies to overwhelm them, along with future perspectives for these promising novel compounds.

The Role of Artificial Intelligence in the Diagnosis and Prognosis of Renal Cell Tumors

The increasing availability of molecular data provided by next-generation sequencing (NGS) techniques is allowing improvement in the possibilities of diagnosis and prognosis in renal cancer. Reliable and accurate predictors based on selected gene panels are urgently needed for better stratification of renal cell carcinoma (RCC) patients in order to define a personalized treatment plan. Artificial intelligence (AI) algorithms are currently in development for this purpose. Here, we reviewed studies that developed predictors based on AI algorithms for diagnosis and prognosis in renal cancer and we compared them with non-AI-based predictors. Comparing study results, it emerges that the AI prediction performance is good and slightly better than non-AI-based ones. However, there have been only minor improvements in AI predictors in terms of accuracy and the area under the receiver operating curve (AUC) over the last decade and the number of genes used had little influence on these indices. Furthermore, we highlight that different studies having the same goal obtain similar performance despite the fact they use different discriminating genes. This is surprising because genes related to the diagnosis or prognosis are expected to be tumor-specific and independent of selection methods and algorithms. The performance of these predictors will be better with the improvement in the learning methods, as the number of cases increases and by using different types of input data (e.g., non-coding RNAs, proteomic and metabolic). This will allow for more precise identification, classification and staging of cancerous lesions which will be less affected by interpathologist variability.