The study was proposed to eligible patients during a medical surveillance consultation with their oncologist. During this consultation, the oncologist provided a full explanation of the study objectives and the protocol. The eligibility criteria were: (1) aged 18–75 years, (2) treated with at least one intensive chemotherapy line, (3) treated for a TGCT or a hematological malignancy [multiple myeloma (MM), non-Hodgkin lymphoma (NLH), Hodgkin lymphoma (HL) or acute myeloid leukemia (AML)], followed or not by an allogenic (allo-HCT) or autologous stem cell transplantation (ASCT), (4) have achieved complete remission according to the reference standard, (5) have completed at least one medical surveillance visit between May 2018 and December 2020, (6) no history of psychiatric disorders, (7) have given written consent to participate. All patients who completed the first study visit were included in the PASCA study population study.

The complications of interest in the study were selected on the basis of two factors: (1) the absence of systematic screening and management within the center, and (2) the limited number of prospective and longitudinal data in the literature.

Following the oncological or hematological follow-up consultation, screening for cardiac, nephrology, pneumology, overweight/obesity, endocrinology and dermatology complications was carried out. The screening lasted about 90 min and included:

Patients attended a second and a third follow-up visit at 6 and 12 months after the first visit, respectively, when similar clilnical and biological investigations were carried out.

The research staff involved were as follows: (1) the investigating physicians who included the patients, (2) a clinical research associate who supervised the completion of the questionnaires, (3) a clinical research nurse who carried out the biological sampling, (4) a physical activity teacher who supervised the physical testing, (5) the study coordinating physician who carried out the medical examination and assessed the complications.

Before the start of the study, the study coordinator physician created a grading system for each complication (Table 1):

The following cut-off values of the French High Authority for Health (HAS) or French or European learned societies were used by the study coordinator physician to detect any abnormal results: cardiology (8, 12, 31), nephrology (32), pneumology (29, 30, 33), overweight/obesity (34), endocrinology (35–37), and dermatology (38). When several guidelines were available, selection was based on the following criteria: availability and applicability of the guideline to the cancer survivor population. When no guidelines were available for cancer survivors, the cut-offs available for the general population were applied. The questionnaires were interpreted in accordance with the recommendations of the authors (30–33).

For a given study visit and for a given complication, if two different grades were identified in a patient, the higher grade was systematically retained. Example of a fictitious patient’s cardiological assessment at his first study visit: hypertension measured with an average of 150/90 mmHg and no known history of hypertension at the time of diagnosis would generate a grade of “Complication.” If he also had dyslipidemia of 1.8 g/L chronically treated with lipid-lowering drugs at the time of diagnosis, it would generate a “Grade 2.” His highest cardiological grade would be “Complication,” and he would therefore be classified in this group for this visit.

The questionnaires were interpreted in accordance with the recommendations of the authors (30–33).

After each visit and for each complication category, the study coordinator physician applied his grading system, and based on the clinical impact for the patient, decided to either provide an oral recommendation to the patient by telephone, refer the patient to their general practitioner or refer them to another healthcare professional.

Patient satisfaction was assessed during the third follow-up visit, using a 4-points Likert scale on four items self-completed by the patients: (1) usefulness of the PASCA program, (2) quality of the information provided by the study team, (3) quality of the orientation provided after the visits, and (4) satisfaction about the length of each visit. The response items were “totally disagree/agree a little bit/agreed somewhat/totally agree” for scales (1), (2), and (3); and “little long/acceptable length/short length/very short length” for scale (4).

We set up a regional network of healthcare professionals specifically for the study, throughout the Auvergne-Rhône-Alpes (AURA) region (8.1 million inhabitants) in order to refer patients, when needed, to a healthcare professional as close as possible to their home and to reduce referral problems. About 45% of the patients attending the Léon Bérard Comprehensive Cancer Center live in Metropolitan Lyon and the Rhône Department (1.9 million inhabitants), and 55% live in neighboring departments (4.5 million inhabitants) or farther away.

For each new patient entering the program, a request to affiliate with the PASCA network was sent by mail or e-mail (with a telephone reminder in case of no response) to the patient’s general practitioner, explaining the objectives of the program and their role in an information leaflet. Positive responses to the request to affiliate were considered as acceptance, and negative responses and non-responses as a refusal.

The primary endpoints of the study were the recruitment rate (number of patients who completed their first visit/number of patients eligible for the study) and the adherence (number of patients who completed all visits/number of patients who completed their first visit). The secondary endpoints were the patient satisfaction and the percentage of de novo complications detected.

Three data extractions were performed by a research technician from:

As the primary outcome of the PASCA study was feasibility, no formal sample size calculation was performed. The sample size in the present study was pragmatic, based on expected recruitment, considering that 90 patients were sufficient to assess the feasibility of the screening procedure. The sample size aligns with published recommendations for feasibility studies (39–42).

Participants’ characteristics were described using median and interquartile range for quantitative data concerning the duration between study timeline (diagnosis, last cancer treatment, first, second and third program visit). Means and standard deviations (SDs) were described for other quantitative data. Frequencies and percentages were used for qualitative data.

Comparative analyses were performed for percentages using Chi-squared or Fisher’s exact tests, and for means using t tests, F-tests or Wilcoxon tests, as appropriate. A type 1 error risk α of 0.05 was used. For multiple comparisons, a Bonferroni correction was applied. A two-sided p value < 0.05 was considered as significant.

A linear mixed model for repeated measures was used to evaluate the relationship of the follow-up (visit number) on HRQoL score, fatigue FSS score and 6 minute walk distance (6MWD). The visit number was considered as a continuous variable. The following assumptions were evaluated: relationship between residuals and distribution of residuals. The best fitted model was selected based on likelihood ratio test. Sankey plots were constructed to graphically represent the proportions of patients graded for complications at each of the three visits, as well as the flow of patients changing grades between visits. For each complication, patients with incomplete data availability at all three visits were excluded from the analyses.

Statistical analyses were performed using the tidyverse, lmerTest and the MuMIn package in R software version 4.1.2.

The baseline demographic and clinical characteristics of the participants are summarized in Table 2. From the 149 patients screened for eligibility, a total of 117 participants were eligible for the present study (Figure 1).

A total of 98 patients consented to participate in the study and completed the first program visit these patients were included in the PASCA study analysis population.

Their mean age (±SD) was 52 (±13) years, with a majority of men (73%), and the mean BMI was 25.1 (±4.6). Overall, 77.6% had hematological malignancies and 22.4% had TGCTs. At inclusion, 74 patients had a first line of anticancer treatment and 24 a second line. The cancer treated was a first relapse or a second cancer for 21 and 4.1% of the patients, respectively. Previously 26% of the patients had undergone surgery, 81% had received only chemotherapy and 19% had received chemotherapy combined with radiotherapy or hormonal therapy. Allo-HCT and ASCT were performed in 12 and 52% of patients, respectively. The medical history at diagnosis of their last cancer included: cardiac (21%), pulmonary (7.8%), endocrine (6.7%), renal (4.4%), and dermatological (2.2%) comorbidities.

The recruitment rate was 83.8%, i.e., a total of 98 patients consented to participate out of the 117 patients who were identified electronically by the research team, prior to the medical consultation.

A total of 42 patients of the initial 98 patients attended the three scheduled study visits, giving an adherence rate of 43%. This subpopulation with longitudinal data was representative of the PASCA study analysis population as shown in Table 2. The reasons for not attending all three visits were: change to a peripheral center closer to the patient’s home, either on the initiative of the oncologist or of the patient for 27 patients (28%); appointment canceled because of the SARS-CoV-2 (COVID-19) pandemic for 12 patients (12%); relapse of multiple myelomas for 13 patients (13%); and consent withdrawal after the second visit for 4 patients (4%).

The percentage of patients with grade ≥ 2 complications (grade 2 or de novo complication) was: 27.6% (overweight/obesity), 13.1% (dermatology), 11.6% (pneumology), 10.2% (cardiology), 5.1% (endocrinology), 2.2% (nephrology). More than 8% of the analysis population (n = 98) with a de novo complication had dermatology, cardiology, and pneumology complications (Supplementary Table S1). No statistical differences were found between patients with TGCTs and hematological malignancies.

There was an increase between the first and third visits in the number of patients who developed a grade ≥ 2 complication (grade 2 or de novo complication) among the 42 patients who completed all three study visits (Figure 2). These complications concerned nephrology (+13.9%), overweight/obesity (+12.5%), endocrinology (+8.3%) and cardiology (5.6%). In contrast, we observed a decrease in dermatology (−10.8) and pneumology (−2.5%) complications between the first and last visits. For each category, the differences between visits were not statistically significant. The evolution observed for complication grades between visits and details of patient flow are available in the Supplementary material.

No relationship was found between the overall HRQoL score and the follow-up (number of visits) (p = 0.11) (Supplementary Figure S1A). We observed a significant (p = 0.003) and negative relationship between the fatigue FSS score and the follow-up (low fatigue FSS score = low intensity fatigue symptoms) (Supplementary Figure S1B), and a significant (p = 0.021) and positive relationship between the 6MWT and the follow-up (high 6MWD = high functional capacity) (Supplementary Figure S1C). For both significant relationships, the conditional R-squared was high (54.6 and 68.7%), while the marginal R-squared (4.5 and 2.1%) was very low, respectively.

Among the 42 patients who attended the three study visits, 34 (81%) completed the satisfaction questionnaire, 23 (68%) patients “totally agreed,” and 10 (30%) “agreed somewhat” with the statement “the PASCA program is helpful to patients”; 19 (56%) “totally agreed” and 14 (42%) “agreed somewhat” with the statement “the information provided by the healthcare staff is clear and understandable.” Finally, for the question “how do you find the length of each PASCA program visit?,” 24 (60%) responded “acceptable length” and 10 (30%) responded “short length.”

Overall, 120 healthcare professionals out of the 150 contacted agreed to join the network, including 97 physicians (54.6% general practitioners and 45.4% specialists), corresponding to a response rate of 80%. However, an average of two telephone reminders were required to achieve this recruitment level. All the healthcare professionals who agreed to join the network, practiced in the Lyon agglomeration.