Chemotherapy Use within the Last 14 Days of Life in Patients with Hematological Malignancies

Hematologic malignancies (HM) are widely variable in their potential for cure, but most have multiple treatment options that are also increasing due to research. The toxicity profile of the treatments can also be quite variable. Thus, discussing realistic therapeutic goals, and potential serious life threatening toxicities, is critical to planning appropriate end of life medical interventions. We analyzed chemotherapy use within 14 days prior to end of life, and do not resuscitate (DNR) orders, among patients with HM who died at our institution.

We reviewed all adult patients (age>18 years) with a diagnosis of HM who had received care at M.D. Anderson Cancer Center and died from 12/1/2010 through 5/31/2012 (pts that had a stem cell transplantation within 100 days of end of life were excluded). Data on patients' demographics, chemotherapy within last 14 days, inpatient DNR orders, and co-morbidities were from the institution's administrative databases. HM diagnostic categories were grouped as Leukemia, Lymphoma, Multiple Myeloma and Other. Statistical analysis: odds ratios (OR) with 95% confidence interval (CI) were compared for different groups (chemotherapy use with age, gender, ethnicity, comorbidities, years of disease diagnosis, and HM diagnostic categories); statistical significance was defined as P value < 0.05.

1,179 HM patients died: median age 65 years; 61% female; 472 (40 %) died in hospital, of which 442 (94%) had DNR orders and 294 (26 %) died in ICU.

Overall, 272 (23 %) had chemotherapy within 14 days of end of life. 29% of pts with leukemia, 19% of pts with lymphoma, 10% of pts with myeloma, and 25%of pts with other diseases received chemotherapy within the last 14 days of life. Forty-nine% (134/272) of them received chemotherapy as frontline therapy; and (219/272 = 81%) died in hospital.

Adjusted by logistical regression, chemotherapy within 14 days of end of life was less common in patients with lymphoma and myeloma than in those with leukemia [odds ratios (ORs), 0.7 and 0.3, respectively; all P<0.05]. It was also less common in older patients (≥80 years) compared to younger patients (<60 years) (OR, 0.5; P<0.05). Chemotherapy within 14 days of end of life was associated with years of disease diagnosis, patients who had disease diagnosis within years 1–3, 3–5, and 5 or more had consistently less treatment prior to death than those patients who were diagnosed within 1 year [ORs, 0.5, 0.5, and 0.3, respectively; all P<0.05]. There were no statistically significant association between chemotherapy within 14 days of end of life and other factors such as genders, ethnicity, and co-morbidities (all P>0.05).

Nearly a fourth of HM deaths at our hospital occurred within 14 days of treatment, and 60% of these were inpatients receiving second line or later therapy. Most pts (94 %) had DNR orders indicating that an end of life decision was discussed, and 26% died in ICU, suggesting acute events such as sepsis likely led to hospital admission. In addition, some patients with HM can manifest serious disease-related complications, and death may also occur shortly after diagnosis. Patients should be informed of these risks for treatment and end of life planning decisions.

No relevant conflicts of interest to declare.