Sebastiani, Silvia

Abstract Background: CINV remains a challenging chemotherapy toxicity, particularly for anthracyclines + cyclophosphamide (AC) for breast cancer. Complete response (CR), defined as no vomiting or use of rescue medication over 5 days after chemotherapy, has been the standard endpoint for successful antiemetic prophylaxis, yet duration of prophylaxis treatment failure (i.e., breakthrough CINV) is rarely assessed. Previous work demonstrated that initial cycle breakthrough CINV is associated with at least one subsequent cycle with CINV (Schwartzberg 2011), and that longer duration CINV is associated with more lost work time and impaired activity (Schwartzberg ASCO 2020). We sought to evaluate the duration of breakthrough CINV in cycle one and its association with individual patients’ repeat breakthrough CINV in subsequent cycles. Methods: Days of CINV was a prespecified endpoint in a prospective, 4-cycle CINV prophylaxis trial of combination netupitant/palonosetron (NEPA) + dexamethasone (Dex) for patients with breast cancer receiving AC; the primary endpoint was CR. Patients without CR were classified as prophylaxis treatment failure (TF) and categorized as short TF (sTF, 1-2 days) or extended TF (xTF, ≥3 days), consistent with work by Ballatori [2006] and Roeland [2019]. We analyzed patients’ sequences of CR, sTF, and xTF for cycles 1-4 and assessed likelihood of CR for cycles 2-4 based on cycle 1 TF duration, using chi-square statistics. Results: Of 402 patients in cycle 1, 303 had CR and 99 (24.6%) had TF. Duration of TF in cycle 1 was sTF for 48 patients while 51 patients had xTF. Patients with sTF in cycle 1 often experienced CR in cycle 2 (32/46 remaining patients; 69.6%) while patients with xTF in cycle 1 often had TF in cycle 2 (38/49; 77.6%). Patients had >84% likelihood of repeating their cycle 2 outcome (CR or TF) in cycles 3 and 4. Over all cycles, those with sTF in cycle 1 had CR in 75/108 later cycles (69.4%) while xTF in cycle 1 led to CR in 32/105 later cycles (29.0%), p<0.001) (see Table). Duration of later-cycle TF was typically consistent with initial cycle TF duration (p=0.046). As a predictor of later cycle TF, xTF had a positive predictive value of 0.695, with sensitivity of 0.689 and specificity of 0.701. Those with xTF in cycle 1 had 2.28 times the relative risk of later TF (CI 1.67-3.11; p<0.001) vs those with sTF. The absolute increase in risk from xTF was 40.4%. Conclusions: Although the majority of patients receiving NEPA + Dex as prophylaxis for AC in breast cancer successfully avoided breakthrough CINV, for patients with prophylaxis treatment failure, CINV duration in cycle 1 strongly predicted their future likelihood of CINV. Patients with extended (≥3 days) breakthrough CINV in cycle 1 had CINV again in over 2/3 of later cycles; conversely those with short breakthrough CINV in cycle 1 avoided CINV entirely in over 2/3 of later cycles. The markedly higher risk faced by patients with initial-cycle extended CINV is both statistically significant and clinically meaningful, due not only to its longer duration and risk-prediction but also to previously demonstrated worsening of work loss and impaired activity. Clinicians should monitor breakthrough CINV duration, seek to avoid extended CINV, and consider changing antiemetic prophylaxis when extended CINV does occur. Further study is needed to confirm this finding in other chemotherapy regimens and with other triple prophylaxis regimens. Impact of Extended CINV in Cycle 1 on CINV in Later CyclesTF1P valueInitial Cycle Result (n/total initial cycles)Total TF1 (99/402)Short TF1 (48/99)Extended TF1 (51/99)Short vs Extended TF1Later CR cycles (n, % of total later cycles)107 (50.2%)75 (69.4%)32 (29.0%)<0.0001Later TF (n, % of total later cycles)106 (49.8%)33 (30.6%)73 (71.0%)Duration of Later TF (n, % of all later TF)Later sTFLater xTFLater sTFLater xTF0.04620 (60.6%)13 (39.4%)29 (39.7%)44 (60.3%)Total later cycles (n, %)213 (100%)108 (100%)105 (100%)TF: prophylaxis Treatment Failure; TF1: TF in cycle 1; CR: Complete Response (0 days TF); CR1: CR in cycle 1; sTF: Short (1-2 days) TF; xTF: Extended (≥3 days) TF Citation Format: Rudolph M. Navari, Gary Binder, Eric J. Roeland, Alexander Molassiotis, Kathryn J. Ruddy, Thomas W. LeBlanc, Dwight D. Kloth, Silvia Sebastiani, Marco Turini, Luke M. Schmerold, Xing Liu, Lee Schwartzberg. Duration of chemotherapy-induced nausea and vomiting (CINV) as a predictor of later-cycle CINV [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS13-27.

Abstract Background: Although many studies have demonstrated consistent levels of effectiveness of CINV prophylaxis for the entire study population across multiple chemotherapy cycles, rarely have studies reported how each patient’s risk of subsequent CINV differs based on prior cycle breakthrough CINV with the same prophylaxis. We lack data on whether prophylaxis continues to fail in the same group of patients each cycle, or whether failure is random with each subsequent cycle. We sought to evaluate individual patients’ risk of repeat CINV in each subsequent chemotherapy cycle. Methods: In a prospective, 4-cycle CINV prophylaxis trial of oral or intravenous combination netupitant/palonosetron (NEPA) + dexamethasone (day 1) for patients with breast cancer receiving anthracycline + cyclophosphamide (AC), we defined CINV as vomiting or use of rescue medication during days 1-5 after chemotherapy. Patients without CINV were classified as complete response (CR); the rest as treatment failure (TF). We analyzed patients’ sequences of CR and TF, and compared CR or TF for cycles 2-4 based on cycle 1 outcomes, using chi-square statistics. To provide context, we performed a post-hoc similar analysis of results reported by Herrstedt et al [2005] from a clinical trial of ondansetron + aprepitant (APR) for patients with breast cancer receiving 4 cycles of AC. Results: The 402 female patients in the NEPA trial received a total of 1,299 cycles. In cycle 1, 99 (24.6%) patients experienced TF (TF1); over all 4 cycles, TF occurred 253 times (19.5%). Patients with CR in cycle 1 (CR1) had a ≥92% rate of CR in cycle 2; their rates of repeat CR were similar in each subsequent cycle. Patients with TF1 had nearly equal risk of CR or TF in cycle 2 (45:55); thereafter 85% of this TF1 subgroup repeated their cycle 2 outcome (CR or TF) in cycles 3 and 4. Over all cycles of NEPA, patients with CR1 subsequently had CR in >90% of cycles 2-4; those with TF1 subsequently had TF in 49.8% of cycles 2-4 (p<0.0001) (see Table). We separately examined Herrstedt’s evaluation of 433 patients across 1537 cycles with APR. In cycle 1, TF was seen 213 times (49.2%), with TF reported in 46.7% of cycles 1-4. We found that patients with CR1 had an 80.5% rate of CR in cycle 2 and repeat CR rates were higher in subsequent cycles. Patients with TF1 had TF in cycle 2 (TF2) at a rate of 78.4%, with 76.6% TF3 and 72.7% TF4. For APR, CR1 resulted in subsequent CR in 78.6% of cycles 2-4 while patients with TF1 again had TF in 74.8% of cycles 2-4 (p<0.0001). Patients with TF1 were more likely to later drop from the study (see Table). Notably, among those with CR1 after APR, the few patients who later had TF in any cycle, had a subsequent repeat failure rate similar to those with TF1. Conclusions: When patients receiving guideline-recommended triple antiemetic prophylaxis successfully avoided CINV in their first cycle of HEC, they had 80-95% likelihood of repeating that success in later cycles. After NEPA, those whose prophylaxis failed in cycle 1 did not face a similar high risk of repeat failure in cycle 2. The pattern of repeat failure after aprepitant was different, with a high repeat failure risk starting in cycle 2. These findings strongly suggest that consistent population average CR rates reported across cycles may mask a higher repeat failure rate for individual patients that experience cycle 1 CINV, particularly for aprepitant. Further study of this phenomenon is needed for other HEC regimens, and to confirm the lack of high repeat failure seen in cycle 2 for NEPA. Repeat CINV in Later Cycles, Based on Cycle 1 ResultsNEPAAprepitant + OndansetronInitial Cycle Result (n/total initial cycles)CR1 (303/402)TF1 (99/402)CR1 vs TF1 (P value)CR1 (220/433)TF1 (213/433)CR1 vs TF1 (P value)Subsequent CR cycles (n, % of total subseq. cycles)636 (93.0%)107 (50.2%)<0.0001464 (78.6%)124 (25.2%)<0.0001Subsequent TF cycles (n, % of total subseq. cycles)48 (7.0%)106 (49.8%)126 (21.4%)370 (74.8%)Total subsequent cycles (n, % of total subseq. cycles)684 (100%)213 (100%)590 (100%)494 (100%)Dropped vs ITT* (n, % ITT cycles)225 (24.8%)84 (28.3%)0.226370 (10.6%)145 (22.7%)<0.0001* The NEPA trial was closed when the last patient completed the first cycle, resulting in an artificially high proportion of patients that did not complete 4 cycles. Citation Format: Rudolph M Navari, Gary Binder, Alexander Molassiotis, Eric Roeland, Kathryn J. Ruddy, Thomas W. LeBlanc, Dwight D. Kloth, Silvia Sebastiani, Lina Y. Dimberg, Luke M. Schmerold, Xing Liu, Lee Schwartzberg. Chemotherapy-induced nausea and vomiting (CINV) risk after prior breakthrough CINV: Unmasking the false average [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS13-09.

Background: Clinician adherence to antiemetic guidelines for preventing chemotherapy-induced nausea and vomiting (CINV) caused by highly emetogenic chemotherapy (HEC) remains poorly characterized. The primary aim of this study was to evaluate individual clinician adherence to HEC antiemetic guidelines. Patients and Methods: A retrospective analysis of patients receiving HEC was conducted using the IBM Watson Explorys Electronic Health Record Database (2012–2018). HEC antiemetic guideline adherence was defined as prescription of triple prophylaxis (neurokinin-1 receptor antagonist [NK1 RA], serotonin type-3 receptor antagonist, dexamethasone) at initiation of cisplatin or anthracycline + cyclophosphamide (AC). Clinicians who prescribed ≥5 HEC courses were included and individual guideline adherence was assessed, noting the number of prescribing clinicians with >90% adherence. Results: A total of 217 clinicians were identified who prescribed 2,543 cisplatin and 1,490 AC courses. Patients (N=4,033) were primarily women (63.3%) and chemotherapy-naïve (92%) with a mean age of 58.6 years. Breast (36%) and thoracic (19%) cancers were the most common tumor types. Guideline adherence rates of >90% were achieved by 35% and 58% of clinicians using cisplatin or AC, respectively. Omission of an NK1 RA was the most common practice of nonadherence. Variation in prophylaxis guideline adherence was considerable for cisplatin (mean, 71%; SD, 29%; coefficient of variation [CV], 0.40) and AC (mean, 84%; SD, 26%; CV, 0.31). Conclusions: Findings showed substantial gaps in clinician adherence to HEC CINV guidelines, including a high variability across clinicians. Clinicians should review their individual clinical practices and ensure adherence to evidence-based CINV guidelines to optimize patient care.

e23108 Background: ASCO antiemetic guidelines recommend upfront triple prophylaxis (NK1 receptor antagonist (RA) + 5HT3 RA + dexamethasone) for patients receiving highly emetogenic chemotherapy (HEC). Physicians exhibit high variation in HEC guideline adherence, principally involving inclusion of NK1 RA. Whether adherence is associated with selection of the specific 5HT3 RA agent remains inadequately studied. Methods: In the IBM Explorys electronic health record database, we used procedure and diagnosis codes to identify HEC courses and related nausea/vomiting (NV) from 2012 to 2018. We defined HEC guideline adherence as triple prophylaxis at chemotherapy initiation. We assigned HEC courses for ≥ 7 day cycles of cisplatin or anthracycline + cyclophosphamide (AC), or carboplatin (≥ 14 day cycles as a proxy for AUC ≥ 4) to prescribing oncologists based on encounter frequency. We categorized 5HT3 RA use of each physician treating ≥ 5 HEC courses based on their most commonly used 5HT3 and performed descriptive statistics. Results: Of 12,262 HEC courses, 57% involved physicians having greater use of ondansetron (OND) (mean OND to palonosetron (PALO) ratio of 3.9:1). These courses had lower guideline adherence (due to NK1 RA omission) and higher rates of NV for combined HEC, AC, and carboplatin. NV rates for cisplatin did not vary by 5HT3 agent used. Courses involving physicians with greater PALO use (mean OND:PALO ratio of 0.2:1) had superior adherence and NV rates, despite involving a slightly higher risk population (younger and/or female). Conclusions: According to HEC antiemetic guidelines, NK1 RA use should occur independent of 5HT3 RA agent selection. However, we observed less NK1 RA use where OND was preferred, which may have caused the observed higher rates of NV with OND. Further evaluation should assess whether pharmacy cost minimization is a driver of both OND preference and NK1 omission in HEC. [Table: see text]

Abstract Background: In order to improve care and reduce costs of cancer treatment, attention has focused on reducing inpatient (IP) and emergency room (ED) utilization. The US Centers for Medicare and Medicaid Services (CMS) recently implemented an oncology outcome measure (OP-35) to assess the quality of care and determine outpatient hospital payment. It assesses 30-day post-chemotherapy rates of IP or ED events deemed “potentially avoidable” by CMS due to association with any of 10 CMS-defined toxicities: anemia, dehydration, diarrhea, fever, nausea, emesis, neutropenia, pain, pneumonia, or sepsis. Although CMS found that 20% of chemotherapy treatment resulted in such events, event rates with highly emetogenic chemotherapy (HEC) are understudied, particularly for breast cancer (bCA) patients treated with anthracycline + cyclophosphamide (AC), carboplatin, or cisplatin. Methods: In a large electronic health record database focused on US integrated delivery networks, HEC courses of therapy were identified from 4Q 2012 to 3Q 2017. IP/ED events within 30 days of chemotherapy administration were considered related to the OP-35 measure if they included a diagnosis code of any of the ten OP-35 toxicities, per CMS' definitions. HEC chemotherapy use, IP/ED events, bCA diagnosis, and toxicities were identified by ICD-9, ICD-10, and procedural codes. Subgroups were evaluated for 3 common HECs: AC, carboplatin (>14 days apart, as a proxy for AUC ≥4), and cisplatin. Results: 4128 courses of HEC in bCA were identified. Of these, 2304 involved AC (median 4 cycles), 1721 involved carboplatin (median 6 cycles), and 103 involved cisplatin (median 4 cycles), with median ages of 55, 59, and 64, respectively. 30-day IP/ED events were seen in 25% of these HEC bCA courses (22%, 30% and 23% for AC, carboplatin, and cisplatin, respectively). The 10 CMS-defined toxicities were associated with an IP/ED event for 73%, 72%, and 76% of events for these AC, carboplatin, and cisplatin courses respectively, confirming that these are principal contributors to 30-day IP/ED use for patients receiving HEC in bCA. The top five that were most commonly associated with IP/ED for bCA for patients receiving AC, carboplatin, or cisplatin were pain 53%/49%/63%, anemia 51%/51%/38%, fever 48%/31%/38%, neutropenia 46%/27%/31%, and nausea/emesis 30%/40%/25%; patients may have had ≥1 toxicity associated with each IP/ED event. The IP/ED rate may be understated because some events, particularly ED, may occur out of the network the EHR data covers. Conclusion: One quarter of patients receiving AC, carboplatin, or cisplatin for bCA visit the ED or are hospitalized within 30 days of at least one of their chemotherapy administrations. Most events involved ≥1 of 10 toxicities deemed avoidable causes of hospitalization by CMS and now tracked as an oncology outcome measure. Relative to patients receiving AC, acute care rates for patients receiving carboplatin were lower for fever and neutropenia and higher for nausea/emesis, perhaps because the study period largely precedes the 2017 addition of carboplatin AUC ≥4 to national guidelines as HEC requiring triplet antiemetic prophylaxis. Toxicity prevention and symptom monitoring are crucial to reduce acute care needs. Citation Format: Ruddy KJ, Roeland EJ, LeBlanc TW, Binder G, Sebastiani S, Potluri R, Schmerold LM, Papademetriou E, Navari RM. Avoidable acute care use associated with nausea and emesis among patients receiving AC, carboplatin, or cisplatin for breast cancer [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P4-16-05.

74 Background: U.S. National Antiemetic Guidelines recommend upfront triple prophylaxis (NK1 receptor antagonist (RA) + 5HT3 RA + dexamethasone) for patients receiving highly emetogenic chemotherapy (HEC), including carboplatin AUC ≥ 4 per 2017 guidelines. While existing data show gaps in guideline compliance, variation between individual physicians is less studied, and a realistic target compliance rate remains unknown. Methods: In a large electronic health record database (IBM Explorys), we identified HEC courses of therapy initiated from 2012 to 2017. Guideline compliance was defined as triple prophylaxis at chemotherapy initiation. Patient courses for ≥ 7 day cycles of cisplatin or anthracycline + cyclophosphamide (AC), or carboplatin (≥ 14 day cycles as a proxy for AUC ≥ 4) were ascribed to oncologists based on encounter frequency. We then ranked physicians treating ≥ 5 HEC courses and evaluated guideline compliance and individual physician variation. Results: In total, 10,074 HEC courses were identified and attributed to 451 unique physicians. Overall antiemetic guideline compliance with cisplatin and AC averaged 68% and 81% respectively. When ranked by compliance, the top 20% of physicians were 2.5 - 1.5 times as compliant as the bottom 20% (cisplatin 100% vs 40%; AC 100% vs 67%). For cisplatin, 32% of physicians had > 90% compliance; the remaining 68% were evenly distributed from 0 - 90%. For AC, 56% of physicians had > 90% compliance, and another 14% had 80 - 90%; the remaining 30% were evenly distributed. For carboplatin, 62% of physicians had ≤ 10% compliance, and another 17% had 11 - 20%; however, the majority of these data preceded guideline inclusion of carboplatin AUC ≥ 4 as HEC. Rates were independent of course volume per physician. Conclusions: Considerable physician-level variation exists in triple antiemetic prophylaxis guideline adherence for HEC. Hundreds of physicians had > 90% compliance with guidelines, suggesting 90% is a realistic target. However, the majority exhibited substantial gaps in NK1 RA use in HEC, placing patients unnecessarily at risk for CINV. Interventions are needed to bolster triple antiemetic prophylaxis in HEC, perhaps especially for carboplatin.