Introduction

Tyrosine Kinase Inhibitors (TKI) have dramatically improved the prognosis of chronic myeloid leukemia (CML) with an expected life expectancy close to the healthy population in newly diagnosed chronic phase CML (CP-CML) patients [1]. While progressionfree survival was the main therapeutic objective of the past two decades, recent recommendations in the field have evolved and quality of life without life-long TKI exposure should be considered [2]. Indeed, numerous clinical trials have now demonstrated that a substantial proportion of patients who have achieved a deep and sustained molecular response can safely discontinue their treatment without molecular relapse defined as major molecular loss [3,4]. Then prolonged treatment exposure associated with the risk of late and sometimes serious adverse events (AE) together with a negative impact on patients’ quality of life can be avoided [5,6]. However, based on recent reports of late molecular relapse, a sustained molecular follow-up after TKI cessation is required [7,8]. Patients must go to their regional specialist medical centre, which is linked to the reference regional laboratory, to consult their doctor and monitor their residual disease. The frequency of follow-up depends on the treatment phase, but varies from every month to every 3 or 6 months. In order to maintain and facilitate this follow-up on a long term basis, we evaluate the feasibility of a remote follow-up based on kits that enable the shipment of blood samples obtained either at the patients’ domicile or neighbourhood laboratory and send by mail to the reference laboratory that carries out the assessment of BCR::ABL1/ABL1^IS ratio and phone or video consultation was performed as soon as the BCR::ABL1^IS result was available. Following the first remote consultation and in order to obtain patients’ feedback a questionnaire was sent to the patients. This study was initially designed for patients who have discontinued TKI treatment in a treatment-free remission (TFR) strategy. Taking into account the rising prevalence of the disease, we decided to extend the study to patients treated with TKI. Here we report a preliminary analysis based on an interim analysis of 224 BCR::ABL1^IS assessments performed in 144 patients over a 21 months period.

Methods

Patients

From March 2020 to December 2021, we included adult patients (≥18 years old), diagnosed in CP-CML. All patients have been informed and have not expressed their opposition to be included in this observational study according to the current French regulations.

Process

Blood samples were collected on 2 ethylenediaminetetraacetic acid (EDTA) blood collection tubes (2*4.5ml) supplied in the kit, at neighbourhood laboratory or patients’ domicile (private home) by the nurse and sent to the reference laboratory with a special mail service using triple packaging required for the transport of organic products. The shipments occurred at ambient temperature with priority mailing. Several kits at different time points of the follow-up could be used in the same patient. BCR::ABL1/ABL1^IS ratio was assessed by reverse transcription quantitative polymerase chain reaction (RT-qPCR) after RNA extraction and reported as ratios of BCR::ABL1 to ABL1 standardized to the international scale (IS) (BCR::ABL1/ABL1^IS) according to previously reported recommendations [8,9]. Three and 2 replicates were performed for BCR::ABL1 and ABL1 genes respectively. A 3 (MR3 or Major Molecular Response, MMR), 4 (MR4) and 4.5 log (MR4.5) molecular responses were defined as a 3-log, 4-log or 4.5-logreduction of transcript, equivalent to BCR::ABL1^IS ≤0.1%, ≤0.01% or ≤0.0032% respectively.

The quality of RNA was evaluated by control gene mean copy number per PCR replicate, i.e. ABL1 gene, and mean number of ABL1 copies per replicate of each sample with a minimum quality criteria of ABL1 ≥ 10000 copies per replicate currently recommended [9,10].

In order to evaluate a potential impact of the delivery time on the quality of RNA, the mean number of ABL1 copies per replicate of each sample was analysed taking into account the time between the date of blood collection to the date of the receipt at the reference laboratory (delivery time).

We also compared for each patient, the difference between the mean of the three previous ABL1 copy numbers per replicate obtained before this remote follow-up and the ABL1 copy number obtained in the study. Differences were analysed according to the delivery time.

As soon as BCR::ABL1^IS result was available, a phone or video consultation according to patient preference was performed. At the end of this remote consultation, medical documents, i.e. medical prescriptions, blood analyses results including BCR::ABL1IS result were generated, archived in patient medical record and if possible shared with the patient using a French software application (Panda lab) [11]. Date and type of the next consultation- remote or at medical center-were established according to patient and physician decisions. If necessary and according to the physician’s point of view, a prompt consultation at the medical center was planned following the video or phone consultation.

To evaluate patients’ feedback on this remote follow-up and to collect specific socio-demographic informations, a questionnaire was sent to the patients following the first video or phone consultation (details of the questionnaire are available in the supplementary appendix).

Statistical Analyses

Socio-demographic variables (Sex, age, employed or retired, internet access and frequency of use, cell phone possession, distance between the medical center and home) and CML patient characteristics (time from CML diagnostic to first remote blood collection sample, molecular response levels and treatment discontinuation phase or TKI treatment at the time of the remote blood analysis) were compared with the level of perception of the video or telephone consultation coupled with sending biologic specimens by mail according to the answer to the question C1 of the questionnaire. Levels of perception were pooled into 3 subgroups: highly prefer and somewhat prefer (group 1), somewhat dislike and very much dislike (group 2) and patients with no preference or no answer (group 3). Socio-demographic variables were compared between the three groups.

All variables were assessed by univariate analysis using Fisher test or T tests. P-value ≤ 0.05 was considered as significant. Age at the date of questionnaire was categorized into 2 groups, with cut-off set at the median. Correlation between ABL1 mean copy number and the delivery time was estimated by calculation of the correlation coefficient R2. Analyses were performed with the SPSS v22 software.

Results

Patients

From March 2020 to December 2021, 144 patients agreed to participate to this remote follow-up. The characteristics of the study population are presented in Table 1. Median age at diagnosis and at the first point of the remote follow-up were 56 and 67 years respectively and 41% were female. At the time of the first blood sample collection, 39 (27.1%) patients were free from TKI in the context of a TFR program with a median time between TKI discontinuation and first remote BCR::ABL1^IS analysis of 4.2 years (range 0.1-12.5).

Blood Samples Collections

Over a 21-month period, 224 BCR::ABL1^ISwere performed in 144 patients using postal kits (one assessment in 87 patients, 2 in 45 patients, 3 in 6 patients, 4 in 2 patients, 5 in 3 patients and 6 in one patient). Six patients have at least 4 BCR::ABL1^IS. Among them, 5 patients underwent BCR::ABL1^ISmonthly evaluation following TKI cessation in a TFR program. Blood samples were obtained at patient domicile or neighbourhood laboratory or both in 36 (25%) patients, 100 (73%) patients and 6 (4.1%) patients respectively.

Quality of RNA

Patient samples arrived at the reference laboratory between 1 and 9 days (median time: 2 days) BCR::ABL1/ABL1^ISquantification failed in 3 out the 224 blood collection samples due to a ABL1 copy number per replicate lower than 10000. This poor RNA quality for these 3 samples was due to either an inadequate preanalytic process (n=2) or a lengthy shipment (9 days for one). Theses 3 patients were contacted and performed another remote blood collection. RNA quality was correlated with the delivery time (R2=0.9326). Thus, the longer the delivery time was, the mean ABL1 copy number per replicate is low (Figure 1-A).

Figure 1A: Number of remote blood samples (kits) analyses according to the xshipment duration and copy number of the control gene ABL1 obtained (n=224remote blood samples).

We observed that a delivery time longer than 4 days led to a significant decrease in the ABL1 copy number per replicate. The impact of the delivery time was also confirmed when difference between the copy number of ABL1 obtained during the remote follow-up and the mean of the 3 last ABL1 copy number per replicate previously obtained for each patient was analysed (Figure 1-B). Patients were then recommended to perform blood collection samples only at the beginning of a given week.

Figure 1B: Difference between the mean of the 3 ABL1 copy numbers previously obtained and the result of the remote analysis for each patient (n=144) according to the delivery time.

Consultations at Medical Center Following Video or Phone Consultations

Following video or phone consultation, a consultation at the medical center was planned in 12/144 patients according to physician decision. Half of these consultations were related to CML management (TKI related AE in 3 patients; TKI resumption after molecular relapse in 3 patients in TFR). The remaining were not related to CML management (worsening of a pre-existing comorbidity in 3 patients, new occurring diseases in 3 patients, all CML and TKI-unrelated).

Patient Sociodemographic Characteristics and Perception of the Remote Follow-Up

Following the first video or phone consultation depending on patient preference, a questionnaire aiming to collect several sociodemographic data and perception of this remote follow-up was sent to the patients. A total of 137 questionnaires were completed and returned. The socio-demographic data of the 137 patients are presented in Table S1. The median age at the date of completed questionnaire was 67 years and 58% of the patients were retired. Eighty-six % of the patients have internet access but only 65% are regular user. Almost all the patients (97%) have a cell phone and 57% were able to download and use applications with the cell phone.

Table S1: Socio-demographic characteristics of the patients who fulfilled and returned the questionnaire (n=137 patients).

Regarding patients’ perception of the advantages of a video or telephone consultation, gain in time and easy organization were the most frequently cited advantages in this multiple response possible question (Figure 2-A). Unsurprisingly, lack of physical examination was the most frequently cited drawback following by difficulties to communicate and to exchange document(s) with the physician (Figure 2-B).

Figure 2A: patient perception of the advantage(s). [Pts: patients].

Figure 2B: The drawback(s). [Pts: patients].

In spite of these reported drawbacks, 80% of the patients were satisfied or very satisfied with these remote consultations (Figure 2-C). Overall, 80% of the patients highly or somewhat prefer this remote follow-up and agreed to use it under normal circumstances, i.e. free from COVID-19 related restrictions (Figure 2-D).

Figure 2C: Video or phone consultation (multiple response possible questions), patient global satisfaction of the video or telephone consultation. [Pts: patients].

Figure 2D: Patient opinion regarding follow-up based on the shipment of local blood samples and video or telephone consultation. [Pts: patients].

In addition, patient preference concerning video or telephone consultation, patient perception of the role of a specialized nurse in the care and follow-up and the wish to receiving validated BCR::ABL1^ISresults have been evaluated (Figure 3, supplementary appendix). Of note, only 36% of the patients were favourable to the delegation of their care and follow-up to a specialized nurse (Figure 3-B, supplementary appendix).

Figure 3A: Patient preference of a video or telephone consultation.

Figure 3B: Patient opinion regarding the role of a specialized nurse in the care and follow-up.

Figure 3C: Patient attempt to receiving validated BCR::ABL1/ABL1IS results.

Factors Associated to the Patients’ Level of Satisfaction of the Remote Follow-Up and Care

Three groups of patients were defined according to the answer to the question C1 of the questionnaire: highly prefer and somewhat prefer (group 1, n=116), somewhat dislike and very much dislike (group 2, n=6) and patients with no preference or no answer (group3, n=15). We failed to identify any factor associated with patients satisfaction level except BCR::ABL1^IS (Table 2).

Table 2: Socio-demographic characteristics of the patients who fulfilled and returned the questionnaire (n=137patients).