Background

The PROCHE [Programme for optimisation of the chemotherapy network] initiative is an innovative oncology-monitoring program designed to reduce patient waiting time and chemotherapy wastage, ultimately improving patient care. A nurse calls patients 48 hours before anticancer treatment at the daily hospital to anticipate chemotherapy preparation.

Methods

Primary objective was to evaluate the incidence of different symptoms reported by grade (NCI-CTC AE: from 0 to 4) prospectively collected from 2008 to 2016. Secondary objective compared the 2009-2016 patients to the control cohort (2008 period) quantified using Mantel-Haenszel khi² and exact p-values.

Results

From January 2009 to December 2016, 3012 patients were enrolled in the program, representing 36 803 questionnaires completed over the whole period. Main adverse events (AE) were collected and compared to the control cohorts (2008, n = 513), resulting in a significant decrease in majority of topics. Global incidence comparison is presented in table 1, with the p-value according to khi2 test. Details of severity levels scores using NCI CTC-AE V4.0, will be reported at ESMO meeting. For some of the toxicities, severe or life threatening toxicities (grades 3-4) were higher in the PROCHE cohort, while global incidence decreased.rnTable:

1542PD

Conclusions

The PROCHE initiative resulted in adverse events decrease and improved patient quality of care and improved hospital as well as pharmacy efficacy.

Legal entity responsible for the study

Florian Scotté

Funding

None

Disclosure

F. Scotté: Roche, Vifor, MSD, Teva, Norgine, Prostrakan, Leo pharma, Janssen, Hospira, Boehringer, Sanofi, Amgen, Pierre Fabre Oncologie, Tesaro. C. Thibault: Roche. All other authors have declared no conflicts of interest.

Background

There is growing concern in society about aggressiveness of cancer care near the end of life (ACCEoL), mainly in metastatic disease. This study aims to determine prevalence and recent time trend of ACCEoL of adult cancer patients in a European country, comparing metastatic with others.

Methods

Cohort study of adults with ICD-9-CM diagnosis of cancer, who died in public hospitals in mainland Portugal (Jan’10 - Dec’15), identified from the Hospital Morbidity database (HMD). HMD provided data on primary cancer site, presence of metastatic disease and primary outcome: a composite ACCEoL indicator aggregating presence of 1 of 14 individual indicators in the last 30 days of life or chemo, immunotherapy or biological agents in the last 14 days of life (expansion of Earle et al 2004 framework). We calculated the prevalence of composite and individual indicators and examined time trends (chi2 test for trend) for the whole cohort, in metastatic disease and for main primary cancers. We considered clinically meaningful > 5% change.

Results

92,155 patients were included (median age 73 yo, IQR 62-81; 61.9% male; 53.0% metastatic). The prevalence of the ACCEoL was 71.1%, 69.9% in metastatic patients vs. 72.6% in others (p < 0.001), and varying by primary cancer from 62.7% in breast to 79.3% in haematological (p < 0.001). The most prevalent individual indicators were > 14 days in hospital (42.7%; 42.3% in metastatic) and surgery (27.8%; 26.4% metastatic). The least prevalent were permanent tracheostomy (0.1%) and percutaneous gastrostomy (0.3%). Primary outcome remained stable overtime and despite some individual indicators showed statistically significant changes in study timeframe, none of these had > 5% change.

Conclusions

Surprisingly, we found unchanged trends of high ACCEoL among adult patients and no clinically meaningful difference for metastatic disease group. A lack of integrated palliative care, even with growing resources in the timeframe analysed, suggest that these have not been enough to reduce ACCEoL. The reduced ACCEoL in patients who died with slow progressive cancers (e.g. breast) suggests that better knowledge of disease trajectories can contribute towards reducing ACCEoL.

Clinical trial identification

N/A

Legal entity responsible for the study

N/A

Funding

Calouste Gulbenkian Foundation, Liga Portuguesa Contra o Cancro - Núcleo Regional do Sul

Disclosure

All authors have declared no conflicts of interest.

Background

There is clear evidence supporting the benefits of early palliative care (PC) for patients with cancer. Despite this, referrals to PC tend to be late and are focused mainly on End of Life Care. This project involves evaluation of a new integrated service between oncology and PC teams. The aim of this service is to normalise early PC alongside active oncology treatment and to ensure that PC is available to all cancer patients with PC needs. This service is based on the introduction of a brief “Triggers” tool to help oncologists accurately identify patients who should be referred to PC. This “Triggers Tool” includes 7 disease and patient-specific indicators of progressive disease and PC need.

Methods

Study population: New patients attending the lung oncology outpatient clinics. Prospective longitudinal observational service evaluation. Objectives: To determine the integrated PC service review rate in eligible patients To describe the PC needs of patients reviewed by Triggers team Patients are scored using the “Triggers” Tool and categorised into “Trigger positive” and “Trigger negative” cohorts. Patient reported outcome measures and assessments of PC need are completed using validated tools (IPOS Integrated Palliative Outcome Scale, Phase, Performance status (PS)). Participants are followed up longitudinally and assessments are repeated at different time points.

Results

In the first 4 months of the new service, 84% (97/115) of eligible patients were reviewed by the Integrated PC service within 2 months of the 1^st outpatient clinic attendance. 75% (73/97) of patients reviewed scored positive on one or more of the Trigger tool items. The PC needs of “Trigger positive” patients, as defined by IPOS, were assessed in 70/73 eligible patients: 97% (68/70) patients had at least moderate PC need on at least one IPOS item. 81% (57/70) had severe or overwhelming PC need. 81% (57/70) of patients were PS 0-1.

Conclusions

Initial data suggest that the Integrated PC service is feasible and identifies patients with PC needs who would benefit from early PC referral. This service evaluation will be continued for 12 months which will facilitate analysis of other patient outcomes e.g. time between PC referral and death.

Clinical trial identification

Not applicable as this is a service evaluation

Legal entity responsible for the study

Royal Marsden Hospital

Funding

None

Disclosure

All authors have declared no conflicts of interest.

Background

Variability of PTX exposure using BSA dosing is well documented and often leads to severe toxicities. While carboplatin is dosed to obtain a specific exposure, paclitaxel is conventionally dosed by BSA, leading to a wide range of exposure. This study compared PTX PK-guided dosing to BSA dosing in a PTX-carboplatin regimen treating stage IIIB/IV NSCLC. This is the final analysis of interim results presented at ASCO 2015 (Poster #375). ClinicalTrials.gov Identifier NCT02058433.

Methods

309 patients with stage IIIB/IV NSCLC were randomized to receive up to 4 cycles of first line 3-weekly carboplatin (AUC 5) and a PTX dose of 175 mg/m² (Arm A), or a PTX PK-guided dose (Arm B) to achieve a time above a PTX plasma concentration of 0.05µM (T_c>0.05) for 26 to 31 hours. Response was classified according to Response Evaluation Criteria in Solid Tumors Group. PTX concentrations were measured by immunoassay; T_c>0.05 was calculated with PK software. The primary endpoint was reduction of grade 4 hematological toxicities.

Results

There were 164 patients in Arm A and 155 patients in Arm B, with 191 males and 128 females participating. PK-guided dose adjustment resulted in doses that were widely distributed 73 – 175 mg/m², and statistically lower than in the BSA arm (by 24%, p < 0.001). Compared to Arm A, PK-guided dosing significantly reduced grade 4 neutropenia by 35% (p = 0.002, 23% vs. 16%) over 4 cycles. The incidence of severe (grade ≥3) neutropenia was also significantly reduced by 25% in Arm B over all cycles (p = <0.001). Additionally, neuropathy (≥ grade 2) was reduced from 20% in Arm A to 8% in Arm B (p = 0.008), representing a 60% reduction over all cycles. Response rates were not significantly different; objective response rates were 23% in Arm A and 29% in Arm B (p = 0.285); stable disease rates were 49% in Arm A and 42% in Arm B (p = 0.0.240).

Conclusions

Results of this study are in agreement with a previous report, and present further evidence that PK-guided dosing reduces severe toxicities. This is accomplished by an overall lowering of dose intensity, while still maintaining efficacy. PK-guided dosing personalizes chemotherapy, and may be useful in patient management.

Clinical trial identification

02058433.

Legal entity responsible for the study

Tonji University Affiliated Shanghai Pulmonary Hospital, Tongji University

Funding

None

Disclosure

All authors have declared no conflicts of interest.

Background

The risk of drug-drug interactions (DDI) increases with the number of comedications. The prognostic impact of DDI in oncology is poorly understood.

Methods

We included 105 patients with advanced NSCLC, 100 patients with advanced ER-negative breast cancer (BC) and 100 hospice inpatients (HO) with advanced malignancies between 2010 and 2015. Data collected included all anticancer and non-anticancer drugs received, age, gender, presence of CNS metastases, smoking status, ECOG performance status (PS), Charlson comorbidity score and overall survival (OS) from the time of incurable cancer. Potential DDI were assessed using the hospINDEX of all drugs approved in Switzerland in combination with the DDI software - flycicle mode (HCI Solutions, Bern, Switzerland). Primary study objective was to assess the prognostic value of the severity of DDI per patient cohort using Kaplan-Meier statistics, uni- and multivariate Cox regression models. The study had a power of 84% to detect a survival difference of 25%.

Results

The median number of drugs was 5 (range 0 to 15) in all patients, lowest in BC (4) and highest in HO (6). A major risk for DDI was detected in 74 patients (24.3%) overall, including 29 NSCLC patients (27.6%), 25 BC patients (25.0%) and 20 HO patients (20%). The number of drugs was significantly associated with the risk of DDI (p < 0.0001). The risk of a major DDI increased from 14% in patients with <4 drugs to 24% in patients with 4-7 drugs, 40% with 8-11 drugs and 67% in patients with >11 drugs. Median OS was 8.6 months in NSCLC, 33 months in BC and 1.2 months in HO. The severity of DDI was significantly associated with inferior OS in BC (HR = 1.34, P = 0.018), but not in NSCLC or HO. The severity of DDI remained significantly associated with OS in BC (HR = 1.34, P = 0.017) after correcting for patient age and ECOG PS.

Conclusions

Severity of DDI is a significant and clinically relevant prognostic factor in advanced BC patients. Prospective trials should evaluate the potential benefit of avoiding polypharmacy in this group of patients. In the meantime, increased caution with polypharmacy seems warranted when treating patients with advanced cancer.

Clinical trial identification

2016-00283 (BASEC, national trial identifier)

Legal entity responsible for the study

Markus Joerger MD-PhD ClinPharm

Funding

None

Disclosure

All authors have declared no conflicts of interest.

Background

Patients treated for cancer are at high risk of drug-drug interactions (DDI), which affects nearly 60% of patients on therapy. We developed a freely available DDI resource (www.cancer-druginteractions.org) to support anti-cancer drug prescribing, based on successful implementation for HIV (www.hiv-druginteractions.org) and hepatitis (www.hep-druginteractions.org) treatments.

Methods

A review of literature and registration documents was performed to evaluate the available evidence for potential DDIs of several oncolytics. Decision trees based on the FDA guideline on DDI studies were used to assess clinical relevance of DDIs. Comedications that are frequently used by cancer patients were selected. Interaction potential of drug combinations was classified using a straightforward ‘traffic light’ classification and quality of evidence was classified using the GRADE system. Advice on management of the interaction was included where appropriate. All records were reviewed by an expert panel of clinical pharmacists/pharmacologists.

Results

Thus far, twelve targeted oncolytics for the indications renal cell, hepatocellular and ovarian carcinoma, gastrointestinal stromal tumors, neuroendocrine tumors and sarcoma have been reviewed. Potential DDIs between oncolytics and > 450 comedications have been classified (Table). Tyrosine kinase inhibitors (TKI) show potential interactions which require action of prescribers in more than 20% of reviewed drug combinations. Monoclonal antibodies (MoAB) show clinically relevant DDIs in only 0.7% of reviewed drug combinations.rnTable:

1544PD Overview of evaluated DDIsc

Conclusions

The DDI checker currently includes comprehensive and ready-to-use advices for DDIs with oncolytics for six indications (these are due to be expanded in the coming months). The freely available, independently developed website with ‘traffic light’ classification will facilitate health care professionals’ and patients’ awareness of potential DDIs between oncolytics and frequently used comedications.

Clinical trial identification

not applicable

Legal entity responsible for the study

Radboud University Medical Center

Funding

Abbvie, Astellas, AstraZeneca, Boehringer Ingelheim, Gilead, Pfizer

Disclosure

K. McAllister, S.H. Khoo: Educational grant from Abbvie and Gilead to perform this project. N.P. Van Erp: Educational grant from Astellas, AstraZeneca, Boehringer Ingelheim, Gilead and Pfizer to perform this project. All other authors have declared no conflicts of interest.

Background

In recent years, immunotherapy, which enhances the immune system's response to tumors, has come into clinical use. At the same time, the use of cannabis, which has potential immune-suppressive effects, is increasing in oncology patients, mainly for palliative indications. A large number of patients is being treated with these two modalities and interaction is possible. The aim of this study was to evaluate the influence of cannabis use during immunotherapy treatment on response rate (RR), progression-free survival (PFS) and overall survival (OS).

Methods

In this retrospective, observational study, data was collected from the files of patients treated with Nivolumab in the years 2015-2016 at Rambam Health Care Campus in Haifa, Israel. Nivolumab was given to 140 patients (89 Nivolumab alone, 51 Nivolumab plus cannabis) with advanced melanoma, non-small-cell lung cancer, and renal clear cell carcinoma. The groups were homogenous regarding demographic and disease characteristics. A comparison of patients treated with Nivolumab plus cannabis to Nivolumab alone was made.

Results

In a multi-variant model, cannabis was the only significant factor which reduced RR to immunotherapy (37.5% RR in Nivolumab alone compared to 15.9% in the Nivolumab plus cannabis group (p = 0.016, OR = 3.13, CI95% 1.24-8.13). Cannabis use was not a significant factor for PFS or OS. Factors affecting PFS were smoking (adj HR = 2.41), brain metastases (adj HR = 2.04), and response to therapy (adj HR = 4.89). Factors that reduced OS were smoking (adj HR = 2.41), brain metastases (adj HR = 2.83), hypertension (adj HR = 2.28), low performance status and disease progression (adj HR = 2.83).

Conclusions

In this retrospective analysis, the use of cannabis in combination with immunotherapy decreased RR to treatment, without affecting PFS or OS. This information can be critical for a large group of patients, and requires caution when starting immunotherapy. Considering the limitations of the study, further prospective clinical study is needed to investigate possible interaction.

Legal entity responsible for the study

Local Helsinki Committee (IRB)

Funding

None

Disclosure

All authors have declared no conflicts of interest.

Background

Evidence is increasing that severe skeletal muscle (SM) loss (sarcopenia) is associated with reduced overall survival (OS) in mCRC pts. We recently found, using data of the randomized phase 3 CAIRO3 study (Lancet, 2015), that SM loss was related to shorter time to progression during first line maintenance treatment (Tx) with capecitabine+bevacizumab (CAP-B) or observation. Subsequently, SM loss during more intensive reinduction Tx by adding oxaliplatin (CAPOX-B) was associated to shorter overall survival (ASCO, 2017). As a potential risk factor for reduced survival we explored whether sarcopenia was associated with dose reductions at start of CAPOX-B reinduction Tx and dose limiting toxicities (DLT) during CAPOX-B reinduction Tx.

Methods

Here, CAIRO3 pts were included who received CAPOX-B reinduction Tx. DLT were defined as any dose delay, reduction, or discontinuation of systemic treatment because of reported CTCAE (v3.0) toxicities at start or during Tx. Poisson regression models adjusted for relevant confounders were used to study the association between sarcopenia and DLT.

Results

A total of 254 pts received CAPOX-B reinduction Tx. 39% of pts were sarcopenic and compared to normal SM pts we found no statistically significant differences in age and sex (sarcopenic vs normal SM: mean age 63.6±9.1 vs 61.9±8.5 yrs, p=.20 and 39% vs 31% females p=.31). BMI was significantly lower in sarcopenic pts, but pts were on average still overweight (25.9±3.8 vs 27.2±3.8 p=.01). Overall, 67% experienced ≥1 DLT. At start of CAPOX-B, 25% had already received a dose reduction and the risk of dose reduction at start was significantly higher for sarcopenic compared to normal SM pts (RR 1.8 95%CI 1.08-2.90). Despite more frequent dose reductions at start, sarcopenic pts did not have a significantly lower risk of DLT during CAPOX-B Tx (RR sarcopenic vs normal SM pts 0.86 95% CI 0.46-1.45).

Conclusions

Sarcopenia was significantly associated with dose reductions at start of CAPOX-B reinduction Tx, and not with DLT during CAPOX-B reinduction Tx. Possible explanations for dose reductions at start might be more frequent toxicities during previous Tx including neuropathy.

Clinical trial identification

NCT00442631

Legal entity responsible for the study

Dutch Colorectal Cancer Group (DCCG)

Funding

Province of Utrecht, The Netherlands

Disclosure

B. Dorresteijn, M. Jourdan: Employee of Nutricia Research All other authors have declared no conflicts of interest.

Background

NEPA, an oral fixed combination of the highly selective NK₁RA netupitant (300 mg) and clinically/pharmacologically distinct 5-HT₃RA palonosetron (PALO, 0.50 mg) is the first fixed-dose antiemetic combination to have been approved. A single oral NEPA capsule plus dexamethasone (DEX) given prior to AC and non-AC highly emetogenic chemotherapy (HEC) showed superior prevention of chemotherapy-induced nausea and vomiting (CINV) over PALO plus DEX for 5 days post-chemotherapy; the safety of NEPA was also well-established in the Phase 2/3 clinical program in 1442 NEPA-treated patients. An intravenous formulation of the NEPA combination (fosnetupitant 260 mg/PALO 0.25 mg) is under development.

Methods

This randomized, multinational, double-blind, stratified (by gender and country) Phase 3 study in chemotherapy-naïve patients with solid tumors was designed to assess the safety of a single 30-minute infusion of IV NEPA prior to initial and repeated cycles of HEC. Patients received either IV NEPA or oral NEPA, both with oral DEX on days 1-4. Safety was assessed primarily by treatment-emergent adverse events (TEAEs) and also by laboratory tests, vital signs and ECGs.

Results

404 patients were included in the safety population (203 IV NEPA, 201 oral NEPA) for a total of 1312 exposures. Overall, 53% of patients were male, 99% were white and the mean age was 60 years. Cisplatin was the most frequent HEC (96% of patients) and lung cancer was most common (55% of patients). The TEAE profiles for cycle 1 (Table) and in all cycles were similar for the two treatment groups. There was no increased incidence of TEAEs in subsequent cycles. No clinically relevant changes in QTc and no cardiac safety concerns were seen. No infusion site reactions related to IV NEPA occurred.rnTable:

1547PD

Conclusions

Intravenous NEPA was shown to be safe and well-tolerated with a similar safety profile to oral NEPA in patients with various solid tumors receiving HEC.

Clinical trial identification

NCT02517021

Legal entity responsible for the study

Helsinn Healthcare, SA

Funding

Helsinn Healthcare SA

Disclosure

L. Schwartzberg: Served as a consultant for Helsinn, Tesaro, Eisai, and Merck and have received research funding from Helsinn and Tesaro. D. Voisin, G. Rizzi: Employee of Helsinn Healthcare. M. Karthaus: Served as and received honoraria for being a consultant for Helsinn and Riemser. All other authors have declared no conflicts of interest.

Background

Suboptimal CINV prevention can negatively impact patients’ (pts) QOL by interfering with daily functioning. Antiemetic guidelines recommend co-administration of a NK₁ receptor antagonist (RA)/5-HT₃RA/corticosteroid to optimize CINV control in pts at high risk for CINV. NEPA, a fixed combination of the NK₁RA netupitant and 5-HT₃RA palonosetron (PALO) has shown superior CINV prevention and improvement in QOL over PALO. A single dose of NEPA recently showed non-inferiority to a 3-day aprepitant/granisetron (APR/GRAN) regimen in preventing CINV in the first head-to-head comparison of NK₁RA-containing regimens. The impact of CINV on pts’ QOL in this study was explored.

Methods

This randomized, double-blind, Phase 3 study in chemotherapy-naïve pts receiving cisplatin-based chemotherapy (CT) assessed the non-inferiority of NEPA versus APR/GRAN for complete response (CR: no emesis/no rescue medication [RM]) rates during the overall (0-120 h) phase post-CT. All pts received dexamethasone on days 1-4. Secondary endpoints included proportion of pts with no emesis, no significant nausea (NSN: <25mm on 100mm VAS), no RM use, and no impact on daily life (NIDL) as assessed by the Functional Living Index—Emesis (FLIE), comprised of vomiting- and nausea-specific questions/domains. The Cochran-Mantel-Haenszel test was used for between group comparisons; non-inferiority testing was not done for secondary endpoints.

Results

Treatment groups were similar for the 828 pts analyzed: male (71%); mean age 55 years; lung cancer (58%). NIDL rates were higher for NEPA, particularly during the delayed phase; similar results were seen for no emesis, NSN, and no RM.rnTable:

1548PD

Conclusions

In this first study comparing NK₁RA regimens, NEPA administered only on day 1 was numerically similar to a 3-day oral APR/GRAN regimen in maintaining functional status in patients receiving highly emetogenic CT.

Legal entity responsible for the study

Helsinn Healthcare, SA

Funding

Helsinn Healthcare

Disclosure

L. Zhang: Consultant for MSD; research funding from MSD and Lilly. S. Lu: Consultant Boehinger and Roche; speaker\'s bureau Lilly; travel expenses from Hutchison and Medipharm Limited. S. Chessari, C. Lanzarotti: Employee: Helsinn Healthcare. K. Jordan: Honorarium/consultant for Helsinn, Tesaro, MDS, and Merck. Travel accommodations from MSD. M. Aapro: Honorarium from Amgen. Consultant for Helsinn, Teva, Hospira, Merck KGaA, Merck, Sandoz, Pierre Fabre, Vifor Pharma, Tesaro. Research funding from Helsinn, Sandoz, Hopsira, Novartis, Pierre Fabre. Expert Testimony for Amgen. All other authors have declared no conflicts of interest.

Background

Chemotherapy-induced peripheral neuropathy (CIPN) is a common side effect of oxaliplatin and taxanes with a negative impact on quality of life (QOL). This study investigates the efficacy of wearing frozen gloves (FGs) during chemotherapy for the prevention of CIPN due to oxaliplatin or taxanes and the influence on patients’ QOL.

Methods

Patients with newly diagnosed cancer starting treatment with oxaliplatin, docetaxel or paclitaxel were eligible for this multicenter randomized controlled trial. Patients were randomized between wearing FGs on both hands during treatment or not wearing FGs. Self-reported CIPN and QOL were measured with the validated EORTC-QLQ CIPN20 and EORCT-QLQ C30 at four time points; baseline (T0), after three cycles (T1), end of chemotherapy (T2) and after 6 months (T3). Subscales were analyzed with analysis of covariance and neuropathy symptoms with logistic regression analysis.

Results

Between February 2013 and May 2016, 180 patients were included, 90 patients in both arms. Thirty-one patients (34%) discontinued the FGs before end of chemotherapy mainly due to discomfort. Intention to treat analyses showed that patients in the FG-group experienced less tingling in fingers/hands at T1 (11% vs. 24%; p=.009) and T2 (28% vs. 43%, p = 0.038) compared to controls. At T3 these differences disappeared (28% vs 24%, p=.0884). FG patients also experienced a trend towards less interference in handling small objects (2% vs 10%, p = 0.06) and opening a bottle (9% vs. 6%, p = 0.06) at T1. FG patients also reported significantly lower motoric problems (mean 8.3 (SD 9.7) vs. 12.8 (SD 13.6), p = 0.013) compared to controls at T1. At T1, those treated with FGs reported statistically significant better QOL on EORTC QLQ-C30 subscales physical (mean 82 vs.74), role (mean 66 vs. 51), cognitive (mean 85 vs. 78), and social functioning (mean 79 vs. 67), and symptom scales fatigue (mean 40 vs. 49) and appetite loss (mean 21 vs. 34), all p < 0.05.

Conclusions

No long-term differences in neuropathy were found, but FGs reduced neuropathy symptoms with better QOL during chemotherapy. Future studies should focus on the biological process of cooling to prevent CIPN.

Clinical trial identification

NL39650.015.12

Legal entity responsible for the study

G. Vreugdenhil

Funding

None

Disclosure

All authors have declared no conflicts of interest.