Abstract: Rasburicase, a recombinant urate oxidase used to manage hyperuricemia in patients at risk of tumor lysis syndrome, is increasingly being used in outpatient settings, but outcomes with this approach have not been extensively documented. We assessed the feasibility of using rasburicase in the outpatient setting using anonymized data from electronic medical records of 400 hospital-based outpatient oncology practices across 42 US states. Information included demographics, cancer type, comorbidity score, use of rasburicase, and outcomes. Outpatient visits of 899 newly diagnosed patients were evaluable, of whom 73 (8%) received rasburicase and 826 (92%) did not. Patients receiving rasburicase were older (65.3 vs 48.1 years), with more emergency (13.7% vs 2.7%) and urgent (13.7% vs 4.2%) hospital admissions, and a higher Elixhauser comorbidity score (1.9 vs 1.4). The proportion of patients with leukemia (53.4% vs 53.3%) and lymphoma (46.6% vs 46.7%) was similar between groups. Despite their increased age and comorbidities, outpatients receiving rasburicase had fewer return outpatient visits and fewer total days at the hospital (inpatient plus outpatient) within 2 and 3 months after their index visit vs outpatients not receiving rasburicase (P < .05). There was no effect on return inpatient visits or total inpatient days. Rasburicase use in the outpatient setting is common and is associated with a decrease in subsequent health care utilization.
Hematologic malignancies account for about 9% of all cancers in high-income countries.1,2 Tumor lysis syndrome (TLS) is a metabolic complication of hematologic cancer therapy that can occur when large numbers of cancer cells are lysed rapidly (either spontaneously or in response to therapy), releasing their contents into the bloodstream. The rapid release of intracellular ions and metabolic by-products leads to the five characteristic features of TLS (hyperuricemia, hyperkalemia, hyperphosphatemia, hypocalcemia, and acute kidney injury), which are accompanied by toxic, and sometimes fatal, clinical effects such as minor loss of kidney function or complete kidney failure, cardiac arrhythmia, seizures, or multi-organ failure. Historically, TLS has been associated with bulky, rapidly proliferating, treatment-responsive cancers (eg, acute leukemias, Burkitt lymphoma); however, the advent of new and highly effective agents with improved antitumor activity against cancer types, such as chronic lymphocytic leukemia and multiple myeloma, suggests that increasing numbers of patients will be at risk of TLS.3
Although consensus exists regarding best-practice management of most aspects of TLS, there is a lack of agreement around the management of hyperuricemia, particularly in relation to whether patients should be managed in a hospital or ambulatory setting. The prevention and treatment of TLS involves the use of medications that decrease uric acid levels (hypouricemic agents), such as allopurinol or rasburicase, as well as electrolyte management and adequate hydration, before and after chemotherapy.4-6 Comprehensive literature searches suggest that, for patients at low and intermediate risk of TLS, allopurinol should be used to manage hyperuricemia, while for those at high risk, it is recommended that rasburicase be administered in the inpatient setting.7,8 Because rasburicase initially was administered only in the inpatient setting, it carried with it administrative and financial considerations for hospital-based outpatient oncology practices.
The overall objectives of this real-world retrospective cohort analysis were to assess the baseline characteristics of patients with lymphoma and leukemia receiving rasburicase vs no rasburicase in the outpatient setting for the management of TLS-related hyperuricemia and to evaluate the impact of rasburicase use vs no rasburicase use on return hospital visits. These data add to the evidence base regarding the feasibility of using rasburicase in the outpatient setting for these patients.
Methods
Patient Population and Study Design
Data from electronic medical records were collected from the US Vizient (formerly MedAssets) database, which includes inpatient and hospital-based outpatient data of patients with leukemia or lymphoma from more than 400 hospitals across 42 US states. Independent, community oncology practices were not included in the analysis. The hospitals consisted of large and small facilities in both urban (87%) and rural (13%) locations.
The use of rasburicase during an outpatient visit to any hospital (defined as an “index visit”) was determined by text-mining the billing file of each patient. Outpatient index visits were defined according to rasburicase use only (with or without rasburicase), as the timing of rasburicase (ie, before, same day, or after anticancer therapy) could not be determined from the outpatient medical records. Procedure coding was used to identify the type of anticancer therapy administered at the index visit (ICD-9, CPT, HCPCS).
The independent variable under investigation was the use of rasburicase during the visit. The following outcome variables, determined by the data set available, were assessed: return inpatient visits (with and without intensive care unit [ICU] usage); return outpatient visits; and total days spent in hospital (inpatient stays plus outpatient visits) post-index visit. Inclusion criteria for this study included: an outpatient visit between January 1, 2010, and June 30, 2016; a diagnosis of leukemia or lymphoma; administration of anticancer therapy at index visit; and diagnosis of TLS or hyperuricemia at index visit.
Statistical Analysis
Logistic regression models were used to test the statistical significance of rasburicase use as a predictor of return visits including return inpatient visits (with and without ICU usage) and return outpatient visits within 1, 2, and 3 months post-index visit. A generalized linear model was used to estimate the total number of days spent in or at hospital within 1, 2, and 3 months post-index visit, based on the use of rasburicase (yes/no). All statistical analyses considered various confounding factors, such as patient characteristics (age, gender, admission type, insurance type, and cancer type), patient comorbidities (the number of Elixhauser comorbidities and Elixhauser score) and hospital characteristics (urban/rural, teaching/nonteaching, number of hospital beds, and region within the United States). Although the two comorbidity scales (Elixhauser and Charlson) captured in the database share many of the same conditions and behave similarly under multivariable analysis, the Elixhauser comorbidity scale was selected for this study due to the greater number of applicable conditions (31 vs 17) and hence higher level of detail.9,10
Results
Baseline Characteristics
Overall, 899 outpatient visits for patients with leukemia or lymphoma were evaluable (Table 1). Patients in the rasburicase group were older compared with patients in the non-rasburicase group (mean age, 65.3 vs 48.1 years). A slightly lower proportion of males was observed in the rasburicase group (63.0% vs 65.9%), along with a much higher proportion of emergency (13.7% vs 2.7%) and urgent (13.7% vs 4.2%) hospital admissions.
Patients receiving rasburicase had a higher mean Elixhauser comorbidity index score compared with patients who did not receive rasburicase (1.9 vs 1.4; Table 2), suggesting that these patients were in a worse state of general health.11 In the rasburicase vs no rasburicase groups, there was a similar proportion of patients with leukemia (53.4% vs 53.3%) and lymphoma (46.6% vs 46.7%), respectively. Patient characteristics, cancers, and outcomes did not differ significantly by year of diagnosis, but statistical power to detect year-over-year differences was limited since, on average, the study only included 138 patients per
year, with an average of only 11 patients per year who received rasburicase.
Hospital Characteristics
In the rasburicase group, all outpatient visits (100%) occurred at urban hospitals, predominantly in southern United States (83.6%); just over half of the visits (60.3%) occurred at non-teaching hospitals (Table 3). In the no rasburicase group, almost all visits (93.5%) occurred at urban hospitals, with about half (56.1%) occurring in southern United States; most visits (71.0%) occurred at teaching hospitals (44.8% at major teaching hospitals, ie, academic medical centers; 26.2% at teaching hospitals that are not academic medical centers but have medical school affiliations). In both the rasburicase and no rasburicase groups, the majority of outpatient visits occurred at large hospitals (> 500 beds; 64.4% and 51.9%, respectively). Additional hospital characteristics shown in Table 4.
Multivariable Analysis
Outpatients with leukemia or lymphoma who received rasburicase spent a similar number of days in hospital as inpatient readmissions as those who did not receive rasburicase (least squares means estimates for total inpatient days within 1, 2, and 3 months post-index visit, respectively: 2.6 vs 1.8, P = .48; 2.9 vs 3.1, P = .94; 3.4 vs 3.7, P = .84). However, outpatients who received rasburicase spent fewer total outpatient days at the hospital within a 1-, 2- or 3-month time frame compared with those who did not receive rasburicase (least squares means estimates for total outpatient days within 1, 2, and 3 months post-index visit: 3.1 vs 5.1, P < .0001; 4.2 vs 7.1, P < .0001; 5.3 vs 8.5 days, P = .0001, respectively). In addition, outpatients who received rasburicase had a lower rate of return outpatient visits to hospital within 2 and 3 months post-index visit compared with those who did not receive rasburicase (odds ratios: 0.48, P = .035; 0.39, P = .008, respectively; for return outpatient visits within 1 month: 0.87, P = .68). Despite this, no significant differences were observed in the rate of return ICU visits or return inpatient visits within 1, 2, or 3 months post-index visit between outpatients who received rasburicase and outpatients who did not receive rasburicase.
Overall, the total number of days spent at the hospital (inpatient stays plus outpatient visits) was lower for the outpatients who received rasburicase compared with those who did not, within 1 month (5.9 vs 7.0 days; P = .17), 2 months (7.7 vs 10.4 days; P = .02), and 3 months (9.5 vs 12.6 days; P = .03) post-index visit. To ensure the robustness of these results, simpler model variations were attempted, in which confounders such as admission type, specific hospital characteristics, and insurance type were removed.
Discussion
Although rasburicase is typically administered in the inpatient setting, this study suggests that patients at risk for TLS can often be managed in an outpatient setting, even when they require rasburicase (Figure 1). The use of rasburicase in an outpatient setting resulted in a lower rate of return outpatient visits and fewer total outpatient days spent at the hospital within 1, 2, and 3 months after the index visit, and fewer total days spent at the hospital (inpatient days plus outpatient days) within a 2- and 3-month time frame, compared with no rasburicase use.
Specifically, outpatients who received rasburicase had a higher mean Elixhauser comorbidity index score, suggesting that these patients were in a worse state of general health compared with outpatients who did not receive rasburicase. Patients receiving rasburicase spent fewer total days at the hospital (inpatient stays plus outpatient visits) within 2 and 3 months post-index visit (P < .05); in addition, they experienced a lower rate of return outpatient visits and fewer total days of outpatient visits over a 1-, 2- or 3-month time frame (the number and length of inpatient hospital visits were similar regardless of rasburicase treatment).
Because rasburicase is usually administered to manage TLS-related hyperuricemia in the inpatient setting, the treatment is associated with an administrative challenge and health-economic considerations for hospital-based oncology practices. Recent research has focused on determining how to administer rasburicase in the most cost-effective manner without sacrificing clinical benefit.12
In 2019, 82,310 cases of lymphoma and 61,780 cases of leukemia are expected to be newly diagnosed in the United States.13 In 2018, the estimated US national expenditure on the health care of people with lymphoma was $14.6 billion (fourth highest overall), while for leukemia, the equivalent expenditure was $6.6 billion (sixth highest).14 Moreover, in the United States, non-Hodgkin lymphoma (NHL) and leukemia are associated with high annual incidence rates (NHL: 18.9 case per 100,000 people; leukemia: 13.8 cases per 100,000 people) and mortality rates (NHL: 5.6 per deaths 100,000 people; leukemia: 6.5 deaths per 100,000 people age-adjusted to the 2000 US standard population).15,16 If rasburicase, which is indicated for patients already receiving anticancer therapy, can be used to successfully manage TLS-related hyperuricemia in the outpatient setting, the need for inpatient hospitalization could be avoided and health care costs reduced.6 Further studies are required to assess the feasibility of using rasburicase in an outpatient setting and the subsequent health-economic and clinical implications.
Limitations
It is difficult to generalize the results of a database analysis such as this. Furthermore, it must be acknowledged that sources of error due to confounding and bias are more common in retrospective than prospective studies (adjusted statistical models were used in this analysis to mitigate against confounding factors such as admission type, hospital characteristics, insurance type, etc.). The timing of rasburicase (before, at the time, or after anticancer therapy) and other treatment details were not captured in the medical records, which is a key limitation. Capturing more information would help us to understand the difference in the cost involved for care of inpatients and outpatients and also the factors which led to deciding whether the patient required inpatient care or outpatient care. To build on this research, future analyses will aim to assess the impact of baseline TLS risk, ethnicity, and hospital type on rasburicase-related outcomes, as well as explore the economic burden associated with the management of hyperuricemia and its sequelae.
Conclusion
This study shows that the use of rasburicase to manage TLS-related hyperuricemia in an outpatient setting may represent a feasible alternative to the currently preferred inpatient approach. The results justify further health-economic evaluation to confirm whether outpatient vs inpatient use of rasburicase is associated with a lower cost burden when administered in hospital-based oncology practices.
References
1. Lichtman MA. Battling the hematological malignancies: the 200 years’ war. Oncologist. 2008;13(2):126-138. doi:10.1634/theoncologist.2007-0228
2. Smith A, Howell D, Patmore R, Jack A, Roman E. Incidence of haematological malignancy by sub-type: a report from the haematological malignancy research network. Br J Cancer. 2011;105(11):1684-1692. doi:10.1038/bjc.2011.450
3. Howard SC, Trifilio S, Gregory TK, Baxter N, McBride A. Tumor lysis syndrome in the era of novel and targeted agents in patients with hematologic malignancies: A systematic review. Ann Hematol. 2016;95(4):563-573. doi:10.1007/s00277-015-2585-7
4. Allopurinol Injection (IV). Drugs.com website. Updated December 1, 2018. Accessed March 2, 2020. https://www.drugs.com/pro/allopurinol-injection.html
5. FDA. Allopurinol (oral). Drugs.com website. Updated July 1, 2019. Accessed March 2, 2020. https://www.drugs.com/pro/allopurinol.html
6. Elitek. Prescribing information. Sanofi-Aventis U.S. LLC.; 2004. Updated September 2017. Accessed March 2, 2020. https://www.accessdata.fda.gov/drugsatfda_docs/label/2017/103946s5101lbl.pdf.
7. Belay Y, Yirdaw K, Enawgaw B. Tumor lysis syndrome in patients with hematological malignancies. J Oncol. 2017;2017:9684909. doi:10.1155/2017/9684909
8. Alakel N, Middeke JM, Schetelig J, Bornhauser M. Prevention and treatment of tumor lysis syndrome, and the efficacy and role of rasburicase. Onco Targets Ther. 2017;10:597-605. doi:10.2147/OTT.S103864
9. Garland A, Fransoo R, Olafson K, et al. The epidemiology and outcomes of critical illness in manitoba. Manitoba Centre for Health Policy (MCHP). April 2012. Accessed March 2, 2020. https://mchp-appserv.cpe.umanitoba.ca/reference/MCHP_ICU_Report_WEB_(20120403).pdf
10. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol. 1992;45(6):613-619. doi:10.1016/0895-4356(92)90133-8
11. Howard SC, Jones DP, Pui CH. The tumor lysis syndrome. N Engl J Med. 2011;364(19):1844-1854. doi:10.1056/NEJMra0904569
12. Dinnel J, Moore BL, Skiver BM, Bose P. Rasburicase in the management of tumor lysis: An evidence-based review of its place in therapy. Core Evid. 2015;10:23-38. doi:10.2147/CE.S54995
13. Leukemia & Lymphoma Society. Facts and statistics. Leukemia & Lymphoma Society. Published 2019. Accessed March 2, 2020. https://www.lls.org/facts-and-statistics/facts-and-statistics-overview/facts-and-statistics.
14. National Cancer Institute. Financial burden of cancer care. Progress Report Cancer. Updated February 2019. Accessed March 2, 2020. https://progressreport.cancer.gov/after/economic_burden.
15. American Cancer Society. Non-hodgkin lymphoma: at a glance. Cancer Statistics Center. Updated 2019. Accessed March 2, 2020. 2019. https://cancerstatisticscenter.cancer.org/#!/cancer-site/Non-Hodgkin%20lymphoma.
16. American Cancer Society. Leukemia: at a glance. Cancer Statistics Center. Updated 2019.
Accessed March 2, 2020. https://cancerstatisticscenter.cancer.org/#!/cancer-site/Leukemia
