Abstract
Introduction
There is growing evidence to support SACT 30-day mortality rate as a key metric for assessing quality of cancer care and a key indicator of avoidable harm to patients from SACT due therapy-related or poor patient selection factors. This has been established for solid organ malignancy. However, data in patients with hematologic malignancy is limited. At the Gloucestershire Hospitals NHS Foundation Trust, a large district general hospital providing specialist care to over 600,000 people, we collated data from our in-house SACT dataset to assess factors that impact the 30-day SACT mortality and establish a benchmark for multidisciplinary, educational mortality meeting discussions on mortality in patients receiving SACT. We aimed to review cases characterizing patients, causes of death and the role of SACT in deaths.
Methods
This retrospective study included all patients aged 18 years or older with a hematologic malignancy who received SACT between Jan 1, 2016, and Mar 31, 2021, irrespective of the number of previous treatment cycles or regimens and irrespective of their response within the disease trajectory. SACT was defined as any cytotoxic chemotherapy, immunotherapies such as monoclonal antibodies, and targeted biological treatments. We calculated 30-day mortality after the most recent cycle of SACT for these patients. We analyzed whether patient, tumor type, and therapy-related factors were associated with the risk of 30-day mortality.
Results
A total of 1903 patients with hematologic malignancy received SACT. The 30-day mortality rate for our center was 3.89% for patients receiving SACT and 4.9% of all-cause mortality. In the 30-day mortality cohort the median patient age at death was 71 years (interquartile range (IQR) 52-86); ratio of men to women, 1.18. 30-day mortality increased with age for patients treated with palliative intent or intent to remission (OR 4.9, 95% CI .985-26.255; p=0.0171). Median duration of 30-mortality post SACT was 17 days (IQR 8-22). The commonest malignancy was plasma cell dyscrasias (54%) and 68% receiving treatments with intent to remission. ECOG performance status was 0-2 at final SACT cycle in 65%, 3 in 27% and 4 in 8%. 47% of patients were receiving their first cycle (median number of cycles received 2; IQR 1-3) and 43% receiving first line therapy (median lines of therapy 2; IQR 1-3). Therapy naïve patients had significantly higher 30-day mortality than those who received previous SACT (OR 3.435, 95% CI 2.06-5.66; p<0·0001). 42% of deaths occurred in hospital and 54% were attributed to primary refractory and/or aggressive disease. Other causes of death included infection (19% neutropenic, 15% non-neutropenic) including COVID-19 (3%), major bleed (5%), thromboembolism (1%) and cardiac event (1%).
Conclusions
SACT data provides "real world" assessment of treatment patterns and outcomes in routine malignant hematology practice. It highlights the importance of systematic data collection and analyses to better inform factors associated with higher risk of 30-day mortality in patients with hematologic malignancy, which ultimately promotes review of clinical decision making and offers an opportunity for service improvements. Our findings shed light into some of the factors that affect the risk of early mortality of patients with hematologic malignancy and provides a way to measure if the health service is improving. The insights into risk factors for 30-day mortality will aid in better understanding for treating clinicians and their patients on predictors of potential harms and benefits of SACT thereby carefully balancing patients' personalized expectations with representative outcomes and treatment risks and should be the focus of discussions about SACT, particularly in pretreated or older patient groups. Neutropenic sepsis remains a significant cause of SACT related mortality warranting further improvements in care. Prospective case discussions together with this retrospective review have generated service improvements at our center including standardizing chemotherapy helpline pathways, systematizing SACT mandatory data recording on e-prescribing system and mandatory response assessments as part of SACT protocols' review. Future research work to examine outcomes based on subtypes of hematologic malignancy and on larger datasets is critical for improving our understanding at delivering the right drugs to the right patients.
No relevant conflicts of interest to declare.
