Cost Saving with Home-Based Chemotherapy Approach in Multiple Myeloma: The ADHOMY Trial

Bortezomib is a rst-line therapy for multiple myeloma and also recommended in many associations for relapsing disease. Hospital-at-Home (HaH) is an attractive alternative to Outpatient Hospital (OH) treatment. The purpose of the study is to compare costs and patient-reported outcomes of two different strategies: exclusive OH-based bortezomib administration or combined administration in both OH and HaH. A prospective non-randomized trial was conducted in Nancy University Hospital. QoL was measured using the EuroQol 5D (EQ-5D) and the European Organization for Research and Treatment of Cancer (EORTC) QoL Questionnaire Core 30 (QLQ-C30). The analysis was conducted from the National Health Insurance System (NHIS) perspective. A total of 42 patients were enrolled in the study. Twenty patients received all bortezomib injections in OH (median of 24 injections) and 22 patients received bortezomib in OH alternately with HaH (median of 28.5 injections, 10.5 in OH and 18.0 in HaH). The average cost per injection was 602.63 € in the OH group versus 479.52 € in the OH/HaH group. Combined administration of bortezomib in OH and HaH was associated with a substantial cost-saving of 20.4% for NHIS and no difference in QoL. This trial responds to chronic patients’ wish to assess innovative ways of care.


Introduction
Cancer prevalence is constantly increasing with 385 000 new cases in 2015 in France but improved management reduced associated mortality (1). Multiple myeloma (MM) accounts for 1% of all cancers and 12% of malignant blood diseases, being the second most frequent hematologic malignancy with a median age at diagnosis of 70 years old. Its rising frequency is linked to the ageing of the population (incidence) and the effectiveness of new molecules on the overall survival (prevalence) (2,3). Bortezomib is a proteasome inhibitor targeting the chymotrypsin-like activity of proteasome 26S which induces cell death by blocking activation cascades in the cancerous cells. The initial Marketing Authorization has been delivered in 2004 in France. The tolerance pro le is mainly marked by hematological toxicity, peripheral (mainly sensitive) neuropathies, arterial hypotension and gastrointestinal disorders.
Bortezomib is administered subcutaneously since 2012, after Moreau et al revealed no inferiority compared to intravenous route and even a decrease in the incidence of peripheral neuropathy (4).
Bortezomib is a standard rst-line therapy and is also recommended in many associations for relapsing disease. For young and t patients, the combination of bortezomib, thalidomide and dexamethasone showed superiority compared to thalidomide and dexamethasone alone in induction therapy prior to autograft in a randomized phase 3 trial (5). In transplant-ineligible patients, the addition of bortezomib to the melphalan prednisone regimen, in rst line treatment, increased progression-free survival (6). Novel agents as anti CD38 monoclonal antibody has also been granted marketing authorization, in combination with lenalidomide and dexamethasone, or with bortezomib and dexamethasone, for the treatment of adults having received at least one prior treatment (7,8). Bortezomib-based treatments are usually prescribed for 4 to 9 cycles with an assessment of myeloma clinical and biological response criteria at the beginning of each cycle. Since September 2016, Nancy Hospital at Home services (HADAN) has handled home administrations bortezomib for patients who lived in their area of activity (up to 25 miles away) and met the National Agency for Accreditation and Health Assessment (ANAES) criteria for chemotherapy at home and (9): absence of severe adaptive or psychological disorders, ability to understand the protocol, absence of cognitive impairment, availability and agreement of the attending physician, home safety and hygiene. The rst cycle of injections is delivered in Outpatient Hospital (OH), as well as the rst day of each following cycle to identify bortezomib adverse events while the remaining injections of each cycle are delivered at home by Hospital at Home (HaH). Patients who did not meet the ANAES criteria or who lived outside the HADAN area of activity got all their injections in OH. HaH may be an interesting alternative to OH exclusive care in order to improve patients comfort during chemotherapy and to avoid overwhelming OH, which carrying capacities are limited. Here we report the results of a cost and QoL analysis comparing hospital and home-based administration of bortezomib in patients with MM.

Inclusion criteria
Patients were considered eligible if they met the following criteria: adults aged 18 years or older, enrolled in a social security scheme, who were to receive a treatment with bortezomib for MM in Nancy University Hospital.
The treatment had previously been approved by a multidisciplinary local team. Patients who were already participating in another trial, had a follow-up and/or treatment for another condition requiring a particular care during the bortezomib treatment period were excluded. In this prospective non-randomized trial, the method of care (i. e. OH or OH/HaH) was based on a routine manner, according to patient's eligibility criteria for HaH (including patient consent, general practitioner consent, home located less than 40 kilometers from Nancy University Hospital).
The study was approved by the ethics committee of Angers II, West of France. Patients gave informed consent including for associated drugs prescribed with bortezomib and were able to declare their method of transport from home to hospital. The ADHOMY trial was registered as NCT03493737 in clinicaltrials.gov.

OH treatment
In OH, patients are welcomed by a nurse who carries out a quick interrogation to identify any objection to the treatment (infectious signs, hemorrhagic syndrome, neuropathy or digestive disorders). Blood pressure, heart and respiratory rates, oxygen saturation and temperature are checked. Before the 1st and 3rd injections, blood tests are performed and patients are given a medical consultation, while the 2nd and 4th sessions are only handled by a nurse, who noti es the medical staff in case of abnormalities. If there is no objection to chemotherapy, the authorized hospital pharmacy is instructed to reconstitute bortezomib injection according to the medical prescription. The chemotherapy protocol is edited on the rst day of each cycle for the whole cure.

Combined administration in OH and HaH
Chemotherapy at home is requested by the referring hospital hematologist on the rst cycle of bortezomib. The HaH coordinating physician and nurse call and/or visit the patient to collect information about his home and social environment. The general practitioner (GP) is informed of the treatment plan as he will be requested to 125 check the patient before the 3rd injection of each cycle, while the coordinating nurse will be in charge of clinical examination before the 2nd and 4th injections. The chemotherapy protocol is edited and sent to HaH on day 1 (D1) for the whole cure. Bortezomib is collected by HaH nurses on the day of the injection, given that the maximum duration between reconstitution and injection must not exceed 8 hours. Chemotherapy at home is administered by registered nurses trained in the care of oncology patients. They are responsible for the proper management of toxic waste, in accordance with the O cial 2006 Regulation report (10). Bortezomib can be administered once or twice a week depending on the associated treatments prescribed. Figure 1 presents the treatment plan in each group: OH and OH combined with HaH.

Outcomes
The primary endpoint was the cost per injection in each group and patients' QoL based on the EQ-5D generic and the QLQ-C30 oncology-speci c questionnaires. They included 5 items with 5 possible answers and 30 items scored from 1 to 4, respectively. Patients were asked to ll the questionnaires on D1 of the rst, second and nal cycle of chemotherapy. The secondary endpoint was reporting of unscheduled hospital admissions, general practitioner and emergency care unit consultations, infections, cytopenia, digestives and neurological disorders related to myeloma and/or the treatment.    EQ5D and QLQC30 range from 0 to 1. Statistics showed no signi cant difference in QOL at baseline, as measured by EQ-5D and QLQ-C30 questionnaires ( Table 3). Patients in OH/HaH group have slightly but non-signi cant higher score of QoL compared to OH group at baseline. After the rst cycle of treatment, patients in OH/HaH group showed also better improvement in QoL compared to OH group. However, at the end of treatment, the gap in QoL between the two groups was narrowed as a consequence of higher improvement in QoL in the OH group. Figure 2 illustrates the sensitivity analysis for the estimated difference in cost and QoL gain. The 95% con dence ellipse is mainly located in the cheaper quadrants on the cost-effectiveness plan.

Discussion
This trial aimed to examine how myeloma patients receiving home-based bortezomib treatment assessed their QoL compared to exclusive hospital-based administration. Results could not evidence any signi cant difference in QoL at the end of treatment. However, we have to emphasize that patient in OH group showed better QoL improvement upon the completion of treatment (+ 0.127) compared to patient in OH/HaH group (+ 0.058). Unbalance in certain characteristics between compared groups at baseline could explain difference in clinical gain. Moreover, patients in OH/HAH were older than those included in OH group (Table1).
This study also showed that a potential cost-saving of 20.4 % per injection can be achieved by NHIS through outsourcing bortezomib administration at home alternatively with OH administration. Transportation costs differ greatly between the two groups because patients treated at home use fewer sanitary transportation and also because transportation costs of HaH nurses are already included in the HGT, unlike the hospital DRG. The cost assessment did not take into account costs related to emergency consultations and hospitalizations or additional blood tests at home as they represented only minimal costs for the NHIS. It did not include also indirect costs related to productivity loss as the ADHOMY population was mostly retired. However, patients' sick leaves and caregivers' time off paid work due to the disease can easily be valued using the human capital approach (13). Petrucci et al. investigated the cost of illness of MM in Italy during one year of disease management, including working days and hours lost by patients and caregivers from a societal perspective (14). The highest work-related productivity losses were reported by patients who underwent allogenic stem cell transplant. In this study, housewives' productivity loss had not been included. From the hospital perspective, outsourcing chemotherapy might induce a decrease in activity for OH as it involves a decrease in the number of patients treated in OH. The creation of an outsourcing package to reward the hospital effort for organizing chemotherapy at home instead of OH administration could counterbalance this loss for hospital activity.
The HaH organization in Nancy involves admitting the patient for two days in HaH (the day before for prechemotherapy evaluation and the day of the injection) instead of only one day for the injection in OH. In HaH, patients cannot be discharged if the period between two hospitalizations is less than 4 days, so if chemotherapy is planned twice a week (on D1, 4, 8 and 11), the patient stays in HaH for the whole cycle and the cost of care might be higher. Discussions are currently being held regarding a speci c cost for pre-chemotherapy assessment days in order to value more accurately each day in HaH. In previous French studies regarding to cost-savings of home bortezomib injection, costs didn't include physician visit and biological tests for one (15) and only direct costs were recorded for the other (16). For both, no QoL evaluation was assessed.
Another interesting result of this study is the median distance from 276 home to hospital (50 miles for OH patients versus 5 miles for OH/HaH patients) which revealed that patients living near big cities bene t more of HaH than countryside living patients. In 2016 in France, only 4% of all chemotherapy sessions were carried out at home, the more active centers being Paris and Lyon (17). Indeed, young and active patients would bene t more from chemotherapy in HaH, especially since they could carry on with their job. But cancer patients getting older over time, it appears essential to develop chemotherapy in HaH in more rural areas to promote home support even far from hospitals.
A large international review on chemotherapy at home was published in 2015.
On the 25 studies analyzed with Drummond criteria, the authors from York and Leeds Universities found little difference for QoL, clinical and psychological effects (18 In the context of MM, health economic evaluation of HaH is of particular relevance. The cost analysis is accurately detailed and shows that QoL can be maintained while saving money for health insurance.
These results might encourage governments to further expand home chemotherapy, especially in the context of Covid-19 pandemia. contributed to the analysis and interpretation of data. All authors revised the work for important content and assume responsibility for data integrity and the decision to submit this manuscript for publication, had full access to the study data, edited and reviewed manuscript drafts and approved the nal version for submission.

Declarations
Ethics approval: The study was approved by the ethics committee of Angers II, West of France. Con dence ellipses for OH/HaH versus OH