5-HT3 Receptor Antagonist Effects in Cancer Patients With Multiple Risk Factors
Palonosetron has been found to be effective in treating chemotherapy-induced nausea and vomiting, but no other studies that have compared it with the other 5-HT3 RAs are currently available.
Chemotherapy-induced nausea and vomiting (CINV) is a common side effect of chemotherapy, and may present in the acute (0-24 hours) or delayed (25-120 hours) phase.1,2 Certain factors such as age <50 years, female gender, prior CINV, anxiety, no or minimal alcohol use, history of motion or morning sickness, and emetogenicity of chemotherapy have been associated with increasing the likelihood of a CINV event.3
The 5-hydroxytryptamine3 receptor antagonists (5-HT3 RAs) have proved highly effective in the prevention of CINV, with current guidelines supporting the use of the 5-HT3 RAs for prophylaxis. The National Comprehensive Cancer Network (NCCN), American Society of Clinical Oncology (ASCO), and Multinational Association of Supportive Care in Cancer/European Society for Medical Oncology (MASCC/ESMO) antiemesis guidelines recommend palonosetron as preferred among 5-HT3 RAs for CINV prophylaxis before moderately emetogenic chemotherapy (MEC). Additionally, the NCCN antiemesis guidelines have granted palonosetron preferred status for CINV prevention with highly emetogenic chemotherapy (HEC) regimens. The MASCC/ESMO guidelines consider palonosetron the preferred 5-HT3 RA for anthracycline plus cyclophosphamide chemotherapy regimens when a neurokinin 1 receptor antagonist is not available.4-6
The effectiveness of palonosetron has also been supported by studies conducted in solid tumors as well as blood cancers, demonstrating that patients receiving palonosetron had significantly lower CINV event rates than patients receiving other 5-HT3 RAs, and that palonosetron can be safely and effectively administered to patients receiving chemotherapy regimens administered over multiple days per cycle.7-10
Despite the proven effect of 5-HT3 RAs in preventing CINV, there has been no published research differentiating these agents in patients with multiple risk factors (higher-risk population). Therefore, we examined the impact of palonosetron versus other 5-HT3 RAs on the incidence of delayed CINV at cycle 1 of chemotherapy based on the presence of certain risk factors in patients with cancer.
A retrospective claims analysis was conducted using the OptumInsight database from December 1, 2005, to June 30, 2011. This database covers millions of lives, with a distribution of 96% of patients commercially insured and 4% of patients covered by Medicaid. Pharmacy and medical claims are available from healthcare providers, facilities, and pharmacies, and information on individual physician visits, medical procedures, hospitalizations, medications, and laboratory tests are available, with charged amounts reported for services rendered.
Cancer patients were initially identified based upon the presence of the first chemotherapy session (HEC, MEC, low emetogenic chemotherapy, and minimally emetogenic chemotherapy) within the specified identification period (December 1, 2005, to June 30, 2011). The first chemotherapy date was defined as the index date. All patients were required to have 12 months of continuous enrollment; information from the 6 months before the index date was used to capture baseline information on patients, and information from the 6 months after the index date was used to assess for CINV or CINV-related utilization. Patients who had a claim for a previous antiemetic 5-HT3 RA or chemotherapy during the pre-index period were excluded, and patients receiving multi-day chemotherapy or who were aged less than 18 years were also excluded. The final population had a claim for a 5-HT3 RA on day 1 of the first chemotherapy cycle and the presence of 3 or more risk factors (age <50 years, previous anxiety, female sex, pre-index vomiting, and highly emetogenic chemotherapy).
The study population was examined for differences in CINV rates by each 5-HT3 RA, stratified by number of risk factors. Because of the nature of claims data, a history of alcohol use could not be appropriately identified. CINV was identified using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes for nausea, vomiting, or related events, and use of rescue medication at any time during the 5 days after chemotherapy. Delayed CINV was defined as a primary or secondary diagnosis of nausea, vomiting, or dehydration (ICD-9-CM codes 787.0, 787.01, 787.02, 787.03, 276.5, 276.51, 276.52) on days 2 to 5 after chemotherapy, or by the use of rescue antiemetic therapy following chemotherapy administration. Rescue medications were identified by J-codes—which are used to identify intravenous (IV) drugs—or National Drug Codes, and consisted of the following medications: olanzapine, promethazine, haloperidol, prochlorperazine, lorazepam, metoclopramide, and/or palonosetron, ondansetron, granisetron, and dolasetron—all of which are administered by IV routes (NCCN guidelines). Any administration of rescue medications, antiemetic medications, or hydration procedures on the day of chemotherapy was not considered to be for delayed CINV.
Descriptive statistics were calculated for continuous and categorical variables at baseline for the study patients. For continuous variables, means (standard deviations) and medians were generated, and for categorical variables, percentages were reported. For initial grouping purposes, study patients were classified into patients who received one of the antiemetic prophylaxes at cycle 1. Demographic and clinical characteristics among the groups were analyzed using the t test for means and the χ2 test for frequencies when applicable. Patient demographics including age, sex, and geographic region were tabulated, and comorbidity burden was assessed using the Charlson Comorbidity Index score, based on a review of medical claims occurring during the 6 months before the index date. Descriptive statistics were also used to summarize the number of patients who had any delayed CINV across all risk factors at cycle 1 without controlling for patient demographic and clinical variables. Among the high-risk patients (3 or more risk factors), logistic regression analysis was performed to calculate the odds ratio (OR) estimates reflecting the likelihood of experiencing an uncontrolled delayed CINV in the palonosetron cohort relative to the other 5-HT3 cohort. SAS software, version 9.2 (SAS Institute, Cary, North Carolina) was used for all data transformations and statistical analyses.
The final analysis included 26,974 patients, with 5651 identified as having 3 or more risk factors—thus considered as being high risk. Overall, patient demographics showed a population that possessed a higher risk of CINV in the palonosetron group, which consisted of individuals that were younger, mostly female, and had higher HEC chemotherapy utilization. Subsequently, the palonosetron group also had higher utilization of aprepitant and dexamethasone compared with the other group. Overall, 15.6% of patients had a CINV event (
). The most common risk factor identified was female sex, followed by HEC chemotherapy (
). Breast cancer was the most common type of cancer in this analysis (n = 10,261).
As seen in
, the percentage of patients experiencing delayed CINV increased as the number of risk factors increased: 13.5%, 14.9%, 15.6%, and 18.4% for 0, 1, 2, and 3 or more risk factors, respectively. High-risk patients (3 or more risk factors) receiving palonosetron had fewer delayed CINV events compared with patients receiving ondansetron, granisetron, or dolasetron (
). ORs versus palonosetron were 1.523 (95% CI, 1.231-1.883), 1.426 (95% CI, 1.146-1.773), and 1.683 (95% CI, 1.271-2.229) for ondansetron, granisetron, and dolasetron, respectively.
Similar to the overall population, a greater percentage of high-risk patients on palonosetron had HEC therapy: 87.9% versus 78.7%, 80.6%, and 78.9% for patients on ondansetron, granisetron, and dolasetron, respectively. Although dexamethasone utilization in the high-risk population was similar in the 2 groups, 13.9% of patients on palonosetron had a claim for aprepitant, versus 6.2%, 3.5%, and 2.1%, for patients on ondansetron, granisetron, and dolasetron, respectively (P <.0001).
The breast cancer population exhibited similar results. All patients in this analysis had at least 1 risk factor present because the very few male breast cancer patients were excluded. Similar to the results shown in Figure 1, the percentage of breast cancer patients experiencing delayed CINV increased as the number of risk factors increased. Additionally, the same effect of palonosetron is witnessed versus the other 5-HT3 RAs. In the breast cancer group, 15.5% of palonosetron patients experienced delayed CINV at cycle 1, versus 21.2%, 20.8%, and 24.7% of patients on ondansetron, granisetron, and dolasetron, respectively. ORs versus palonosetron were 1.468 (95% CI, 1.118- 1.928), 1.433 (95% CI, 1.090-1.884), and 1.795 (95% CI, 1.269-2.539), respectively. Similar to the overall population, the palonosetron group in this analysis also had a higher utilization of HEC chemotherapy. Dexamethasone utilization was similar between groups; however, 14.8% of patients in the palonosetron group had a claim for aprepitant, versus 6.8%, 3.9%, and 1.1% in the other groups, respectively (P <.0001).
In terms of the overarching effect of risk factors on CINV, this analysis seems to be consistent with other published studies. Osoba and colleagues3 reported an increased incidence of post chemotherapy vomiting (PCV) between those having no risk factors (20% incidence) and those having 4 of the 6 identified risk factors in the study (76% incidence). Furthermore, there was an approximately 30% difference in the incidence of nausea in patients having any 6 of the 7 potential risk factors (96.2% vs 66.7%, respectively).3 As seen in our study and the analysis done by Osoba and colleagues, there is a potential effect as risk factors accumulate. This could dictate and reinforce the importance of not only assessing emetogenicity of chemotherapy, but also certain patient demographics when assessing CINV risk.
Another study, by Feinberg and colleagues,11 examined incidence and independent risk-factors for CINV during cycle 1. This analysis confirmed chemotherapy type, younger age, and female sex as being independently associated with an increased CINV risk. In addition, Sekine and colleagues12 further examined the effect of CINV risk factors in individual phases of CINV (acute and delayed) as well as the effect of multiple risk factors, and concluded that the number of risk factors was significantly associated with CINV in both phases. Independently, only female sex was associated with treatment failure in both acute and delayed phases. Nonhabitual alcohol intake, age, and nonsmoking status were associated with treatment failure only in the acute setting.12
To our knowledge, this is the first analysis that looks at the differences between the available 5-HT3 RAs for patients with multiple risk factors present. Although we can further corroborate the known association between risk factors and impact on CINV, additional research is needed to evaluate the benefits of using available pharmacologic agents to prevent CINV in high-risk cancer patients.
Several limitations associated with this analysis merit disclosure. Because this was a claims analysis, it is possible that all cases of CINV were not captured; however, severe cases of CINV were likely captured, because patients would have incurred resource utilization for treatment, prompting a medical claim. Moreover, drug therapy was also evaluated for rescue medications utilizing ICD-9-CM codes and National Drug Codes, which would signal a CINV event. Because the analysis was based on medical claims only, it was not possible to capture humanistic implications, such as quality of life.
Furthermore, only effects in cycle 1 were examined, and outcomes in high-risk patients in subsequent cycles of therapy were not collected. Additionally, more patients received aprepitant in the palonosetron group. This difference may be because of a higher proportion of HEC-treated patients utilizing more aggressive antiemetic prophylaxis; NCCN guidelines recommend the addition of aprepitant to a 5-HT3 RA for HEC to properly control delayed CINV. Lastly, other risk factors, such as previous alcohol use or motion sickness, were not collected because of the nature of the data and population. Such risk factors require retrospective chart analysis or a prospective study.
In this analysis, the probability of a delayed CINV event increased with the accumulation of identified risk factors. Palonosetron was associated with fewer delayed CINV events compared with other 5-HT3 RAs in high-risk cancer patients. Further research examining all the associated risk factors and their effects on CINV outcomes is warranted.