Antiemetic guideline consistency and incidence of chemotherapy-induced nausea and vomiting in US community oncology practice: INSPIRE Study.

PURPOSE: Consensus guidelines for preventing chemotherapy-induced nausea and vomiting (CINV) are variably implemented in practice. The purpose of this study was to evaluate the impact of guideline-consistent/guideline-inconsistent CINV prophylaxis (GCCP/GICP) on the incidence of no CINV after cycle 1 of highly or moderately emetogenic chemotherapy (HEC or MEC). PATIENTS AND METHODS: This prospective observational study enrolled chemotherapy-naive adult outpatients who received single-day HEC or MEC at four oncology practice networks, all using electronic health record (EHR) systems, in Georgia, Tennessee, and Florida. Results from the Multinational Association of Supportive Care in Cancer Antiemesis Tool, a validated tool to measure CINV, administered 5 to 8 days postchemotherapy, were merged with EHR data. The primary end point, no CINV, defined as no emesis and no clinically significant nausea (score < 3 on 0-10 scale), was compared between cohorts using logistic regression. RESULTS: A total of 1,295 patients were enrolled (mean age, 59.3 years; 70.0% female; 35.5% HEC). The overall prevalence of GCCP was 57.3%. When corticosteroids were prescribed on days 2 to 4 after all HEC, GCCP for HEC increased from 28.7% to 89.8%; when NK1-receptor antagonists were prescribed after all MEC, GCCP for MEC increased from 73.1% to 97.8%. Over 5 days postchemotherapy, the incidence of no CINV was significantly higher in the GCCP cohort than the GICP cohort (53.4% v 43.8%; P < .001). The adjusted odds of no CINV with GCCP was 1.31 (95% CI, 1.07 to 1.69; P = .037). CONCLUSION: Increased adherence to antiemetic guidelines could significantly reduce the incidence of CINV after HEC and MEC.

Assessment of treatment and monitoring patterns and subsequent outcomes among patients with chronic myeloid leukemia treated with imatinib in a community setting.

INTRODUCTION: Real-world treatment and monitoring patterns have not been well documented among imatinib-treated chronic phase chronic myeloid leukemia (CP-CML) patients. Thus, we evaluated these patterns and responses to imatinib in CP-CML patients. METHODS: This retrospective study, based on the Georgia Cancer Specialists' electronic medical record (EMR) system, identified CP-CML patients initiating treatment with imatinib from 01/01/2002 to 11/01/2011 who were subsequently followed for ≥6 months. RESULTS: A total of 177 patients met the study criteria. Imatinib dose modification occurred in 59 patients (33%). Rates of treatment interruption, discontinuation, and switching to another therapy were 16%, 24%, and 23%, respectively. Of 27 patients discontinuing imatinib for lack of efficacy, 9 (33%) had initial dose escalation; 26 patients (96%) eventually switched to a second-generation tyrosine kinase inhibitor. By 3 months, 168 patients remained on imatinib, of whom 96 (57%) had undergone cytogenetic and/or molecular testing. The frequency of response monitoring fluctuated over time, with rates as high as 28% for cytogenetic and 69% for molecular testing. Cumulative response rates steadily increased; 18 month rates were 47% for complete cytogenetic response and 26% for major or complete molecular response. There were no cases of progression and/or death among 38 patients who were regularly monitored for molecular response within the first 12 months of imatinib. Ten of 98 patients (10%) not regularly monitored had progressed or died. CONCLUSIONS: Almost one-third of patients initiating imatinib for CP-CML required dose modification, treatment interruption, or discontinuation. Opportunities for improved monitoring in this setting were identified. Limitations include those inherent to retrospective analyses based on EMR and the uncertain extrapolability of the results.

Hemoglobin trends and anemia treatment resulting from concomitant chemotherapy in community oncology clinics.

PURPOSE: This study describes changes in hemoglobin (Hb) levels in patients during chemotherapy before anemia treatment and in those who received no treatment, measuring time from chemotherapy initiation to Hb less than 10 g/dL and anemia treatment initiation. PATIENTS AND METHODS: This retrospective longitudinal cohort study used a database of outpatient oncology practice electronic medical records from January 1, 2006, to August 6, 2009. Unit of analysis was episode of chemotherapy care, beginning at initiation and including consecutive administrations within the next 60 days. Patients received two or more administrations of conventional chemotherapeutic agents, had a cancer diagnosis with at least 60 days of follow-up, and had no myelodysplastic syndrome. A total of 4,864 episodes (4,021 patients) met selection criteria, 73% with baseline Hb of 11 g/dL or greater and 60% receiving no anemia treatment. RESULTS: Episodes without anemia treatment increased from 44.6% (2006) to 77.8% (2009). Erythropoiesis-stimulating agent (ESA) use decreased from 45.4% (2006) to 11.5% (2009). Patients receiving transfusions increased from 3.4% to 8.7% (2006 to 2009; all P < .001). Total proportion of episodes with Hb less than 10 g/dL at anemia treatment increased from 16.2% to 93.1% during the same period (P < .001). Mean Hb values before anemia treatment decreased over time from 10.8 to 8.9 g/dL (2006 to 2009; P < .001). Overall, time from chemotherapy initiation to first anemia treatment increased from 24.7 to 36.9 days (2006 to 2009; P < .01). CONCLUSION: Results suggest increased restrictions were associated with decreased use of ESAs and increased use of transfusions as well as delays in anemia treatment and lower Hb levels before anemia treatment. Additional investigation of the overall impact of delayed treatments on long-term patient outcomes is warranted.

Impact of initiating antiemetic prophylaxis with palonosetron versus ondansetron on risk of uncontrolled chemotherapy-induced nausea and vomiting in patients with lung cancer receiving multi-day chemotherapy.

PURPOSE: The purpose of this study is to examine the risk of uncontrolled chemotherapy-induced nausea/vomiting (CINV) among lung cancer patients receiving multi-day chemotherapy and ondansetron- or palonosetron-initiated prophylactic antiemetic regimens in a community oncology setting. METHODS: The Georgia Cancer Specialists electronic medical records database was used to retrospectively identify lung cancer patients who received multi-day cisplatin or carboplatin regimens with ondansetron or palonosetron on day 1 between April 1, 2006 and July 31, 2009. Uncontrolled CINV events were identified through ICD-9-CM codes (nausea/vomiting), CPT codes (dehydration), rescue medications, nausea/vomiting hospitalizations, and/or antiemetic therapy after last chemotherapy administration of the cycle. Risk for uncontrolled CINV, up to 7 days after last chemotherapy administration, was analyzed at cycle level using logistic regression with regressors of gender, age, number of chemotherapy administration days, Charlson comorbidity index, cancer type, multicancer diagnoses, and chemotherapy regimen. RESULTS: A total of 209 palonosetron and 153 ondansetron patients (702 and 515 cycles, respectively) met the inclusion criteria. Palonosetron patients were significantly older (mean 67.9 versus 63.9 years; P < 0.0001), with no significant difference in gender, baseline comorbidity score, or multicancer diagnosis. Palonosetron cycles had 63% lower risk for uncontrolled CINV events versus ondansetron cycles [odds ratio (OR) 0.37; 95% confidence interval (CI) 0.25-0.54; P < 0.0001]. Sub-analysis by chemotherapy supported overall analysis (cisplatin OR 0.09; 95% CI 0.04-0.25; P < 0.0001; carboplatin OR 0.46; 95% CI 0.30-0.70; P = 0.0003). CONCLUSION: In this retrospective analysis of lung cancer patients, multi-day chemotherapy cycles administered with palonosetron on day 1 were associated with a significantly lower risk for uncontrolled CINV events versus ondansetron-initiated chemotherapy cycles.

Hematologic outcomes of myelodysplastic syndromes treatment with hypomethylating agents in community practice.

INTRODUCTION: Hypomethylating agents (HMAs) treat myelodysplastic syndromes (MDS) through suppression of abnormal clones that may cause low hemoglobin (Hgb), platelet (PLT) deficiencies, and reduced absolute neutrophil count (ANC). Our study examined hematologic outcomes in MDS among patients treated with HMAs in a large community hematology-oncology practice. MATERIALS AND METHODS: A retrospective study using electronic medical record data studied patients who received at least one cycle of a single HMA (decitabine [DAC] or azacitidine [AZA]) for MDS from June 1, 2006, to May 31, 2009, who had pretreatment and end-of-treatment Hgb, PLT counts, and ANC available. Multivariate logistic regression assessed predictors of end-of-treatment response (Hgb ≥ 11 g/dL without transfusion or erythrocyte stimulating agent; PLT ≥ 100,000 cells/µL without transfusion; ANC ≥ 1000 cells/mm(3) without colony stimulating factor) adjusting for baseline laboratory values, age, gender, and comorbidities. HMA choice was studied as a predictor of outcome. RESULTS: A total of 137 patients (mean age, 72.2 years; 57% male) met full inclusion criteria (DAC = 84, AZA = 53). Mean number of cycles was four (range, 1-16 cycles) for DAC and five (range, 1-23 cycles) for AZA. Total number of cycles significantly predicted Hgb, PLT, and ANC response (odds ratio [OR] 1.19, P = .029; OR 1.15, P = .031; OR 1.16, P = .047, respectively). Growth factor use at any point during HMA treatment was negatively associated with Hgb and ANC response (OR 0.85, P = .007; OR 0.96, P = .046). There was no difference between treatments in likelihood of PLT or ANC response. CONCLUSIONS: Patients treated with HMAs for MDS are more likely to achieve hematologic response when treated with a greater number of cycles.

Likelihood of a subsequent chemotherapy-induced nausea and vomiting (CINV) event in patients receiving low, moderately or highly emetogenic chemotherapy (LEC/MEC/HEC).

OBJECTIVE: To assess the likelihood of subsequent chemotherapy-induced nausea and vomiting (CINV) events following a first chemotherapy administration CINV event in patients receiving single-day low, moderately, or highly emetogenic chemotherapy (LEC, MEC, or HEC). METHODS: A retrospective analysis was conducted utilizing Georgia Cancer Specialists, Florida Cancer Specialists, and ACORN electronic medical records databases (April 2006 through July 2009). Patients were included who received more than one single-day LEC, MEC, or HEC administration (oral or intravenous) with no chemotherapy 3 months prior to the first LEC, MEC, or HEC administration. Two cohorts, patients with a first administration CINV and no first administration CINV, were created and followed for 6 months. A multivariate logistic regression assessed the likelihood of subsequent CINV, controlling for age, gender, Charlson comorbidity index, cancer type, number of chemotherapy administrations, gap between LEC, MEC, or HEC administrations, and number of different LEC, MEC, or HEC agents administered. RESULTS: A total of 10,586 patients met the inclusion criteria (LEC = 3099; MEC = 5172; HEC = 2315). Of those patients, 4.4% (n = 136), 7.8% (n = 402), and 13.8% (n = 320) experienced a CINV event with their initial single-day LEC, MEC, or HEC administration, respectively. The unadjusted subsequent CINV rate was higher in the cohorts with first LEC, MEC, or HEC administration CINV for all groups receiving LEC (33.1% vs. 16.0%; p < 0.0001), MEC (46.5% vs. 18.9%; p < 0.0001), or HEC (59.1% vs. 26.9%; p < 0.0001). After controlling for covariates, patients with first LEC, MEC, or HEC administration CINV were 3.1, 3.8, and 3.7 times more likely to have a subsequent CINV compared to patients without a first LEC, MEC, or HEC administration CINV (Odds Ratio: 3.05 [95% CI: 2.08-4.48, p < 0.0001]; 3.77 [95% CI: 3.04-4.68, p < 0.0001]; and 3.70 [95% CI: 2.88-4.74, p < 0.0001], respectively). CONCLUSION: In this retrospective analysis, patients receiving single-day LEC, MEC, or HEC who had a prior CINV were at increased risk of subsequent CINV. Further studies assessing increased risk of a subsequent CINV events are warranted given this study represents an assessment of electronic medical record data within select community-based populations under usual care.