PRognostic factor of Early Death In phase II Trials or the end of 'sufficient life expectancy' as an inclusion criterion? (PREDIT model)

Three-month life expectancy as inclusion criterion for clinical trials in advanced pancreatic cancer: is it really a valid tool for patient selection?

PURPOSE

To analyze the 3-month life expectancy rate in pancreatic cancer (PC) patients treated within prospective trials from the German AIO study group.

PATIENTS AND METHODS

A pooled analysis was conducted for patients with advanced PC that were treated within five phase II/III studies conducted between 1997 and 2017 (Gem/Cis, Ro96, RC57, ACCEPT, RASH). The primary goal for the current report was to identify the actual 3-month survival rate, a standard inclusion criterion in oncology trials.

RESULTS

Overall, 912 patients were included, 83% had metastatic and 17% locally advanced PC; the estimated median overall survival (OS) was 7.1 months. Twenty-one percent of the participants survived < 3 months, with a range from 26% in RC57 to 15% in RASH. Significant predictors for not reaching 3-month OS were > 1 previous treatment line (p < 0.001) and performance status (p < 0.001).

CONCLUSIONS

Despite the definition of a life expectancy of > 3 months as a standard inclusion criterion in clinical trials for advanced PC, a significant proportion of study patients does not survive > 3 months.

TRIAL REGISTRATION NUMBERS

NCT00440167 (AIO-PK0104), NCT01729481 (RASH), NCT01728818 (ACCEPT).

Association of pharmacologic thromboprophylaxis with clinically relevant bleeding and hospital-acquired anemia in medical inpatients: the risk stratification for hospital-acquired venous thromboembolism in medical patients study

BACKGROUND

Pharmacologic thromboprophylaxis (pTPX) might exacerbate the risk of clinically relevant bleeding (CRB) and hospital-acquired anemia (HAA) in older multimorbid inpatients.

OBJECTIVES

We aimed to evaluate the association of pTPX use with CRB and HAA.

METHODS

We used data from a prospective cohort study conducted in 3 Swiss university hospitals. Adult patients admitted to internal medicine wards with no therapeutic anticoagulation were included. pTPX use was ascertained during hospitalization. Outcomes were in-hospital CRB and HAA. We calculated incidence rates by status of pTPX. We assessed the association of pTPX with CRB using survival analysis and with HAA using logistic regression, adjusted for infection, length of stay, and the International Medical Prevention Registry on Venous Thromboembolism bleeding risk score.

RESULTS

Among 1305 patients (mean age, 63.7 years; 44% women, 90% at low risk of bleeding), 809 (62%) received pTPX. The incidence of CRB was 2.4 per 1000 patient-days and was not significantly higher in patients with pTPX than in those without. We found no significant association between pTPX and CRB. HAA was frequent (20.2%) and higher in patients with pTPX than in those without (23.2% vs 15.3%). The incidence of HAA was 21.2 per 1000 patient-days and did not significantly differ between patients with pTPX and those without. We found an association between pTPX and HAA (adjusted odds ratio, 1.4; 95% CI, 1.0-2.1).

CONCLUSION

Our study confirmed the safety of pTPX in medical inpatients at low risk of bleeding but identified an association between pTPX and HAA. Adherence to guidelines that recommend administering pTPX to medical inpatients at increased venous thromboembolism risk and low bleeding risk is necessary.