Development of a Clinical Prediction Model for 1-Year Mortality in Patients With Advanced Cancer.

Importance: To optimize palliative care in patients with cancer who are in their last year of life, timely and accurate prognostication is needed. However, available instruments for prognostication, such as the surprise question ("Would I be surprised if this patient died in the next year?") and various prediction models using clinical variables, are not well validated or lack discriminative ability. Objective: To develop and validate a prediction model to calculate the 1-year risk of death among patients with advanced cancer. Design, Setting, and Participants: This multicenter prospective prognostic study was performed in the general oncology inpatient and outpatient clinics of 6 hospitals in the Netherlands. A total of 867 patients were enrolled between June 2 and November 22, 2017, and followed up for 1 year. The primary analyses were performed from October 9 to 25, 2019, with the most recent analyses performed from June 19 to 22, 2022. Cox proportional hazards regression analysis was used to develop a prediction model including 3 categories of candidate predictors: clinician responses to the surprise question, patient clinical characteristics, and patient laboratory values. Data on race and ethnicity were not collected because most patients were expected to be of White race and Dutch ethnicity, and race and ethnicity were not considered as prognostic factors. The models' discriminative ability was assessed using internal-external validation by study hospital and measured using the C statistic. Patients 18 years and older with locally advanced or metastatic cancer were eligible. Patients with hematologic cancer were excluded. Main Outcomes and Measures: The risk of death by 1 year. Results: Among 867 patients, the median age was 66 years (IQR, 56-72 years), and 411 individuals (47.4%) were male. The 1-year mortality rate was 41.6% (361 patients). Three prediction models with increasing complexity were developed: (1) a simple model including the surprise question, (2) a clinical model including the surprise question and clinical characteristics (age, cancer type prognosis, visceral metastases, brain metastases, Eastern Cooperative Oncology Group performance status, weight loss, pain, and dyspnea), and (3) an extended model including the surprise question, clinical characteristics, and laboratory values (hemoglobin, C-reactive protein, and serum albumin). The pooled C statistic was 0.69 (95% CI, 0.67-0.71) for the simple model, 0.76 (95% CI, 0.73-0.78) for the clinical model, and 0.78 (95% CI, 0.76-0.80) for the extended model. A nomogram and web-based calculator were developed to support clinicians in adequately caring for patients with advanced cancer. Conclusions and Relevance: In this study, a prediction model including the surprise question, clinical characteristics, and laboratory values had better discriminative ability in predicting death among patients with advanced cancer than models including the surprise question, clinical characteristics, or laboratory values alone. The nomogram and web-based calculator developed for this study can be used by clinicians to identify patients who may benefit from palliative care and advance care planning. Further exploration of the feasibility and external validity of the model is needed.

Discussing prognosis and the end of life with patients with advanced cancer or COPD: A qualitative study.

OBJECTIVES: To explore patients' experiences and recommendations for discussions about their prognosis and end of life with their physicians. METHODS: Patients with advanced cancer or advanced chronic obstructive pulmonary disease (COPD) were enrolled in qualitative interviews, which were analyzed with a phenomenological and thematic approach. RESULTS: During interviews with fourteen patients (median age 64 years), we identified the following themes for discussion about prognosis and the end of life: topics discussed, the timing, the setting, physician-patient relationship, responsibilities for clinicians, and recommendations. Patients preferred the physician to initiate such discussion, but wanted to decide about its continuation and content. The discussions were facilitated by an established physician-patient relationship or attendance of relatives. Patients with cancer had had discussions about prognosis at rather clear-cut moments of deterioration than patients with COPD. Patients with COPD did not consider end-of-life discussions a responsibility of the pulmonologist. Patients recommended an understandable message, involvement of relatives or other clinicians, sufficient time, and sensitive non-verbal communication. CONCLUSIONS: Patients appreciated open, sensitive, and negotiable discussions about prognosis and the end of life. PRACTICE IMPLICATIONS: Patients' recommendations could be used for communication training. Possible differences in the need for such discussions between patients with cancer or COPD warrant further research.

Predictors of mortality in chronic obstructive pulmonary disease: a systematic review and meta-analysis.

BACKGROUND: Better insight in patients' prognosis can help physicians to timely initiate advance care planning (ACP) discussions with patients with chronic obstructive pulmonary disease (COPD). We aimed to identify predictors of mortality. METHODS: We systematically searched databases Embase, PubMed, MEDLINE, Web of Science, and Cochrane Central in April 2020. Papers reporting on predictors or prognostic models for mortality at 3 months and up to 24 months were assessed on risk-of-bias. We performed a meta-analysis with a fixed or random-effects model, and evaluated the discriminative ability of multivariable prognostic models. RESULTS: We included 42 studies (49-418,251 patients); 18 studies were included in the meta-analysis. Significant predictors of mortality within 3-24 months in the random-effects model were: previous hospitalization for acute exacerbation (hazard ratio [HR] 1.97; 95% confidence interval [CI] 1.32-2.95), hospital readmission within 30 days (HR 5.01; 95% CI 2.16-11.63), cardiovascular comorbidity (HR 1.89; 95% CI 1.25-2.87), age (HR 1.48; 95% CI 1.38-1.59), male sex (HR 1.68; 95% CI 1.38-1.59), and long-term oxygen therapy (HR 1.74; 95% CI 1.10-2.73). Nineteen previously developed multicomponent prognostic models, as examined in 11 studies, mostly had moderate discriminate ability. CONCLUSION: Identified predictors of mortality may aid physicians in selecting COPD patients who may benefit from ACP. However, better discriminative ability of prognostic models or development of a new prognostic model is needed for further large-scale implementation. REGISTRATION: PROSPERO (CRD42016038494), https://www.crd.york.ac.uk/prospero/ .

The development of the ADO-SQ model to predict 1-year mortality in patients with COPD.

BACKGROUND: Goals of end-of-life care must be adapted to the needs of patients with chronic obstructive pulmonary disease (COPD) who are in the last phase of life. However, identification of those patients is limited by moderate performances of existing prognostic models and by limited validation of the often-recommended surprise question. AIM: To develop a clinical prediction model to predict 1-year mortality in patients with COPD. DESIGN: Prospective study using logistic regression to develop a model in two steps: (1) external validation of the ADO, BODEX, or CODEX models (A = age; B = body mass index; C = comorbidity; D = dyspnea; EX = exacerbations; O = airflow obstruction); (2) updating of best performing model and extending it with the surprise question. Discriminative performance of the new model was assessed using internal-external validation and measured with area under the curve (AUC). A nomogram and web application were developed. SETTINGS/PARTICIPANTS: Patients with COPD from five hospitals (September-November 2017). RESULTS: Of the 358 included patients (median age 69.5 years, 50% male), 63 (17%) died within a year. The ADO index (AUC 0.73) had the best discriminative ability compared to the BODEX (AUC 0.71) or CODEX (AUC 0.68), and was extended with the surprise question. The resulting ADO-surprise question (SQ) model had an AUC of 0.79. CONCLUSION: The ADO-SQ model offers improved discriminative performance for predicting 1-year mortality compared to the surprise question, ADO, BODEX, or CODEX. A user-friendly nomogram and web application (https://dnieboer.shinyapps.io/copd) were developed. Further external validation of the ADO-SQ in patient groups is needed.

Predictors of Mortality in Patients with Advanced Cancer-A Systematic Review and Meta-Analysis.

To timely initiate advance care planning in patients with advanced cancer, physicians should identify patients with limited life expectancy. We aimed to identify predictors of mortality. To identify the relevant literature, we searched Embase, MEDLINE, Cochrane Central, Web of Science, and PubMed databases between January 2000-April 2020. Identified studies were assessed on risk-of-bias with a modified QUIPS tool. The main outcomes were predictors and prediction models of mortality within a period of 3-24 months. We included predictors that were studied in ≥2 cancer types in a meta-analysis using a fixed or random-effects model and summarized the discriminative ability of models. We included 68 studies (ranging from 42 to 66,112 patients), of which 24 were low risk-of-bias, and 39 were included in the meta-analysis. Using a fixed-effects model, the predictors of mortality were: the surprise question, performance status, cognitive impairment, (sub)cutaneous metastases, body mass index, comorbidity, serum albumin, and hemoglobin. Using a random-effects model, predictors were: disease stage IV (hazard ratio [HR] 7.58; 95% confidence interval [CI] 4.00-14.36), lung cancer (HR 2.51; 95% CI 1.24-5.06), ECOG performance status 1+ (HR 2.03; 95% CI 1.44-2.86) and 2+ (HR 4.06; 95% CI 2.36-6.98), age (HR 1.20; 95% CI 1.05-1.38), male sex (HR 1.24; 95% CI 1.14-1.36), and Charlson comorbidity score 3+ (HR 1.60; 95% CI 1.11-2.32). Thirteen studies reported on prediction models consisting of different sets of predictors with mostly moderate discriminative ability. To conclude, we identified reasonably accurate non-tumor specific predictors of mortality. Those predictors could guide in developing a more accurate prediction model and in selecting patients for advance care planning.

The prognostic value of the 12-, 6-, 3- and 1-month 'Surprise Question' in cancer patients: A prospective cohort study in three hospitals.

OBJECTIVE: This prospective study aimed to evaluate the performance of the 'Surprise Question' (SQ) 'Would I be surprised if this patient died in the next 12 months?' in predicting survival of 12, 6, 3 and 1 month(s), respectively, in hospitalised patients with cancer. METHODS: In three hospitals, physicians were asked to answer SQs for 12/6/3/1 month(s) for inpatients with cancer. Sensitivity, specificity, positive and negative predictive values were calculated. RESULTS: A total of 783 patients were included, of whom 51% died in the 12-month period after inclusion. Sensitivity of the SQ predicting death within 12 months was 0.79, specificity was 0.66, the positive predictive value was 0.71 and the negative predictive value was 0.75. When the SQ concerned a shorter survival period, sensitivities and positive predictive values decreased, whereas specificities and negative predictive values increased. In multivariable logistic regression analysis, the SQ was significantly associated with mortality (OR 3.93, 95% CI 2.70-5.71, p < 0.01). CONCLUSIONS: The 12-month SQ predicts death in patients with cancer admitted to the hospital reasonably well. Shortening the timeframe decreases sensitivities and increases specificities. The four surprise questions may help to identify patients for whom palliative care is indicated.

The Value of the Surprise Question to Predict One-Year Mortality in Idiopathic Pulmonary Fibrosis: A Prospective Cohort Study.

BACKGROUND: Idiopathic pulmonary fibrosis (IPF) is a progressive fatal disease with a heterogeneous disease course. Timely initiation of palliative care is often lacking. The surprise question "Would you be surprised if this patient died within the next year?" is increasingly used as a clinical prognostic tool in chronic diseases but has never been evaluated in IPF. OBJECTIVE: We aimed to evaluate the predictive value of the surprise question for 1-year mortality in IPF. METHODS: In this prospective cohort study, clinicians answered the surprise question for each included patient. Clinical parameters and mortality data were collected. The sensitivity, specificity, accuracy, negative, and positive predictive value of the surprise question with regard to 1-year mortality were calculated. Multivariable logistic regression analysis was performed to evaluate which factors were associated with mortality. In addition, discriminative performance of the surprise question was assessed using the C-statistic. RESULTS: In total, 140 patients were included. One-year all-cause mortality was 20% (n = 28). Clinicians identified patients with a survival of <1 year with a sensitivity of 68%, a specificity of 82%, an accuracy of 79%, a positive predictive value of 49%, and a negative predictive value of 91%. The surprise question significantly predicted 1-year mortality in a multivariable model (OR 3.69; 95% CI 1.24-11.02; p = 0.019). The C-statistic of the surprise question to predict mortality was 0.75 (95% CI 0.66-0.85). CONCLUSIONS: The answer on the surprise question can accurately predict 1-year mortality in IPF. Hence, this simple tool may enable timely focus on palliative care for patients with IPF.

Physicians' views on the usefulness and feasibility of identifying and disclosing patients' last phase of life: a focus group study.

OBJECTIVES: Accurate assessment that a patient is in the last phase of life is a prerequisite for timely initiation of palliative care in patients with a life-limiting disease, such as advanced cancer or advanced organ failure. Several palliative care quality standards recommend the surprise question (SQ) to identify those patients. Little is known about physicians' views on identifying and disclosing the last phase of life of patients with different illness trajectories. METHODS: Data from two focus groups were analysed using thematic analysis with a phenomenological approach. RESULTS: Fifteen medical specialists and general practitioners participated. Participants thought prediction of patients' last phase of life, i.e. expected death within 1 year, is important. They seemed to find that prediction is more difficult in patients with advanced organ failure compared with cancer. The SQ was considered a useful prognostic tool; its use is facilitated by its simplicity but hampered by its subjective character. The medical specialist was considered mainly responsible for prognosticating and gradually disclosing the last phase. Participants' reluctance to such disclosure was related to uncertainty around prognostication, concerns about depriving patients of hope, affecting the physician-patient relationship, or a lack of time or availability of palliative care services. CONCLUSIONS: Physicians consider the assessment of patients' last phase of life important and support use of the SQ in patients with different illness trajectories. However, barriers in disclosing expected death are prognostic uncertainty, possible deprivation of hope, physician-patient relationship, and lack of time or palliative care services. Future studies should examine patients' preferences for those discussions.

Characteristics and outcome of pediatric non-Hodgkin lymphoma patients with ovarian infiltration at presentation.

BACKGROUND: Ovarian infiltration in pediatric non-Hodgkin lymphoma (NHL) at presentation is rare and information on outcome is scarce and mainly based on case reports and small series. PROCEDURE: Evaluation of clinical characteristics and outcome of ovarian infiltrated pediatric NHL cases of a single center, and an extensive review of the all cases reported so far in literature. RESULTS: At presentation, 6/60 female NHL cases of our center had ovarian infiltration, and combining these cases with earlier case reports, a total of 42 cases were identified. Median age at presentation was 10.9 years (range 0-18), and all but one had a B-cell immunophenotype, with 32/42 cases being classified as Burkitt. Bilateral involvement was reported in 26/41 cases, of which 22 were bilaterally ovariectomized as first treatment. All cases were treated with chemotherapy. Relapses were reported in 9/36 and death in 16/36. After follow-up in our center (median 13.4 years), in 2 cases anti-Müllerian hormone (AMH) values were available (2.1 and 0.9 µg/L), in non-ovarian cases median 2.2 µg/L. CONCLUSIONS: We conclude that in case of ovarian tumors with negative markers, NHL should be considered in order to avoid unnecessary surgery.