The objective of this study was to investigate the perceptions of Brazilian ALL patients about diagnosis, treatment and adverse events and to analyze associations between these variables and patient demographics, disease status and risk of adverse events.

This cross-sectional study included all patients with ALL registered in the Brazilian Lymphoma and Leukemia Association (ABRALE) until 2019. ABRALE is a non-governmental civil society organization (NGO) that supports patients with leukemia and lymphoma and their families by offering educational programs, publications and scientific/interactive events in Brazil. Patients can contact ABRALE and be registered in the database in different ways. ABRALE representatives in treating centers approach some patients while others contact ABRALE to get information, guidance or help on healthcare issues.

This study protocol was approved by the Teaching and Research Ethics Committee of the Albert Einstein Institute. This study was supported and sponsored by ABRALE and AMIGOH (Friends from Oncology and Hematology). Informed consent was obtained and recorded by telephone from all participants during interviews. Data are treated with confidentiality and only anonymous aggregated data are reported here.

For this study, all patients diagnosed with ALL and registered with ABRALE were recruited for this study in December 2019. At least three attempts were made to contact patients. All were interviewed by telephone about their experience of the care they were receiving for ALL. Alternatively, caregivers were interviewed, which was generally the case for under 17-year-old patients. Patients were excluded if contact information on file was out-of-date (out-of-reach) or if they did not consent to participate in the study. The flowchart summarizes how many patients were eligible and enrolled in this study. Enrolled patients were treated from 1981 until January 2019 and registered in ABRALE database (Figure 1).

Figure 1.

Patients eligible and enrolled in this study.

(0.16MB).

The majority of patients were from nine large Brazilian cities (Rio de Janeiro, Curitiba, Goiânia, Recife, Porto Alegre, Belo Horizonte, São Paulo, Campinas and Salvador) or from smaller cities in the states of Amazonas, Ceará, Espírito Santo, Maranhão, Mato Grosso, Mato Grosso do Sul, Pará, Paraíba, Pernambuco, Santa Catarina, Rondônia and Tocantins.

A specific questionnaire was designed for this study to collect data related to patient demographics, the healthcare system, disease status, treatment management and related difficulties, satisfaction and adverse events. Some of the questions were adapted from the quality-of-life instrument of the European Organization for Research and Treatment of Cancer (EORTC QLQ-C30).9 Interviews were conducted in Portuguese (Table 1).

Telephone interviews were conducted by a multi-disciplinary team that had received training and was qualified for this task with knowledge about ALL and good communications skills.

If the first attempts were unsuccessful, at least three home or mobile calls were made to contact interviewees within the optimal timeframe. Patient demographics, basic symptoms, diagnosis and treatment data were assessed during the first successful contact. The interviews lasted about 15–20 min. The responses were input into the ABRALE electronic database.

The questionnaire (Table 1) consisted of open field and multiple-choice questions, depending on the variables of interest. The basic sociodemographic variables were gender, age at interview, age at ALL diagnosis, presenting symptoms, time to diagnosis, state/city of treatment and residence, relationship with caregiver (when applicable) and healthcare system used (private or public). There were also questions relating to treatment protocols and duration, use of cranial radiotherapy, treatment difficulties and satisfaction, minimal residual disease (MRD) status, allogeneic hematopoietic stem cell transplantation (allo-HSCT) and donor type. From the patient perspective, information regarding difficulties to proceed to an allo-HSCT and treatment related to adverse events were recorded. Secondary variables were dichotomizations or regrouping of recorded primary variables, as required for statistical analysis.

The categorical variables are reported as frequencies/proportions. The numeric variables are reported as means, medians, ranges and 95 % confidence intervals (95 % CIs) according to distribution. Potential univariate associations between categorical variables and indication of allo-HSCT or development of adverse events were explored using the chi-square or Fischer exact tests depending on sample distribution. For statistical comparisons, patients were grouped according to the healthcare system accessed (private versus public), while patients that reported access to both health systems (mixed healthcare) were excluded from this comparative analysis. Moreover, patients were grouped and analyzed according to allo-HSCT status, MRD and incidence of adverse effects.

Variables with p-values <0.1 were selected for a multivariate analysis using the logistic regression method.

The Shapiro-Wilk method was applied to test for normality. The Student t-test was used to compare parametric variables, while the Wilcoxon-Mann-Whitney (unpaired) test was used to compare non-parametric data. Modified Poisson regression or negative binomial regression models were chosen, according to variable dispersion, to investigate risk factors associated with the cumulative number of adverse events with results being reported as incidence rate ratios (IRRs).10

The STATA v.11 and R v.3.5.2 computer programs were used for statistical analysis. The level of significance was set for p-values <0.05

Four hundred patients were enrolled in this study. Patient characteristics were male 203 (50.7 %) and the mean age 15.7 (range: 0–77) years with 152 (38 %) being over 20 years old. The questionnaires were completed by the patient and not by a caregiver in 128 (32 %) of the cases. Among the remaining 272 cases, the caregiver who responded the interview was one of the patient's parents in 242 (89 %) of the cases. Patients were diagnosed with ALL from 1981 to 2019, while the median date of diagnosis was September 2015 (interquartile range [IQR]: March 2007-February 2018). The median follow-up after ALL diagnosis until this survey was 6.2 years (range: 0.9–38.6).

Most patients 304 (76 %) were from the southeastern or southern regions of Brazil and were treated in the public healthcare system (280 - 70 %), while 79 (20 %) received treatments from private/medical insurance healthcare systems. Forty-one patients (10 %) reported that their treatments were from ‘mixed’ sources, using both the public and private healthcare systems.

The most commonly reported symptoms at presentation were fever (39 %), bleeding/ecchymosis (38 %), intense fatigue (30 %) and musculoskeletal pain (28 %) (Figure 2). The proportion of patients diagnosed within one week after symptoms onset differed between public (17.9 %) and private healthcare systems (31.1 % - p-value = 0.019). Moreover, diagnostic difficulties were greater for public healthcare patients (35 % versus 22.6 %; p-value = 0.034). The main diagnostic problems reported by patients were referrals to multiple institutions (65/134 - 48.7 %), repeated medical visits (80/134 - 59.8 %), inconclusive diagnostic test results (35/134 - 26 %), inaccessibility to essential laboratory tests (17/134 - 13 %) and difficulty to be admitted in a tertiary hospital (4/134 - 3 %).

Figure 2.

Frequency of symptoms of acute lymphoblastic leukemia patients.

(0.27MB).

All but two patients were treated from 1992 to January 2019 (one in 1981 and one in 1986) with a median follow-up from diagnosis to study survey date of 6.2 years (range: 0.9–38.6). The great majority of patients (91 %; 364/400) did not know their treatment protocol. Thus, only 36 (9 %) patients knew their treatment protocol; overall, patients were able to mention eight protocols (Table 2); two were reported most, the GBTLI (10 - 27.8 %) and the Berlin-Frankfurt-Münster (BFM in 8 - 22.2 %) protocols.

In general, 70 patients (17.5 %) required modifications of treatment attributed to the following reported causes: 32 (45.7 %) upfront refractoriness to chemotherapy, 26 (37.1 %) severe adverse events and three (4.3 %) relapses. Four patients (5.7 %) modified their treatment because they moved to another treatment center and five (7.1 %) did not know why there was a modification.

Forty-eight patients (12 %) had at least one treatment difficulty: 26 (54.1 %) reported that a necessary drug was not available at the time, 12 (25 %) reported the need to buy medications themselves and ten (20.8 %) had to petition to demand drugs to start their indicated treatment. Minimal residual disease (MRD) status was remembered by 99 (24 %) of the patients; it was positive in 22 (23.9 %) cases.

More than half of patients 206 (51.5 %) reported a treatment duration of two years or more, while the treatment lasted six months or less for 87 (21.7 %); among these patients 27.6 % (24/87) were submitted in first remission to allo-HSCT, justifying the shortened treatment protocol.

Cranial radiotherapy was reported by 109 patients (27.2 %) as part of their treatments. Interestingly, comparing the group of patients reporting cranial irradiation versus those treated without cranial irradiation, univariate analysis showed that a higher frequency of cranial irradiation was associated with increased age (>15 years old: 35.0 % versus 19.1 %; p-value = 0.005) and indication of allo-HSCT (51.7 % versus 20.1 %; p-value <0.001). While being diagnosed after September 2015 was associated with less irradiation (21.0 % versus 33.1 %; p-value = 0.01). In multivariate analysis, indication of allo-HSCT remained associated with a higher risk (odds ratio [OR]: 4.60; 95 %CI [95 % confidence interval]: 2.69–7.79; p-value <0.001) of cranial irradiation. In contrast, being diagnosed after September 2015 was associated with a lower risk (OR: 0.57; 95 % CI: 0.35–0.92; p-value = 0.021) of cranial irradiation and age was not significant in this multivariate analysis

In this study, 91 patients (22.7 %) reported referral for allo-HSCT by their physician. At the time of the study survey, 64 (70.3 %) had already undergone allo-HSCT. Among these patients, 31 (48 %) had received grafts from matched related donors, 24 (38 %) from matched unrelated donors, six (9 %) from haploidentical donors and three (5 %) had receiving umbilical cord grafts. Among the patients not submitted to transplants, 19 (20.9 %) were still waiting for donors for the procedure and eight (8.8 %) were waiting for hospital admission.

The transplantation centers were located mainly in the southeastern and southern regions of Brazil. Most patients 34 (53.1 %) needed to move to another city to proceed with the allo-HSCT, while 16 (25.0 %) needed to move far beyond, from one state to another, and 30 (46.9 %) underwent allo-HSCT in the same city as they were living. Forty patients (62 %) reported that more than three months elapsed between the indication for allo-HSCT and the transplant procedure; in 27 (42 %) the waiting time was longer than six months. Seventeen patients (27 %) reported difficulties with the allo-HSCT procedure: for 11 (65 %), the delay was more than three months to find a donor and arrange donor-related procedures, and two patients reported a long delay for hospital admission due to bed/room unavailability. No significant difference was found on analyzing the transplant delay of more than three months in respect to the donor type: matched related 13/31 (42 %) versus matched unrelated (7/24; 29 %; p-value = 0.24).

The following factors were associated with a higher probability of indications for allo-HSCT: MRD positivity (OR: 3.22; 95 %CI: 1.07–9.66; p-value = 0.037), cranial radiotherapy (OR: 4.36; 95 %CI: 1.61–11.8; p-value = 0.004) and age (OR: 1.03/year; 95 %CI: 0.99–1.06; p-value = 0.068).

Overall, 390 (97.5 %) patients reported satisfaction with their treatments. The remaining 10 (2.5 %) patients were unsatisfied due the following reasons: long-term adverse events, insufficient attention to their issues by healthcare professionals or inability to resume social and working relationships.

Despite the high satisfaction rate, 167 (41.7 %) reported at least one ongoing treatment-related adverse event. The most frequently reported adverse events were mood disorders (44/167 - 26.3 %), neurologic impairment characterized by sensorial/strength deficiencies (23 - 13.8 %) or cognitive/memory disorders (20 - 12.0 %), visual problems (18 - 10.8 %), including cataracts, glaucoma and low visual acuity, lung disease (25 - 15 %), such as chronic obstructive/restrictive diseases and emphysema, and liver disease (18 - 10.8 %), such as fatty liver, chronic hepatitis and cirrhosis. The reported frequencies of clinically-relevant adverse events are listed in (Figure 3).

Figure 3.

Frequency of adverse events reported by acute lymphoblastic leukemia patients. HAS: Systemic arterial hypertension; GVHD: Graft-versus-host disease.

(0.22MB).

The risk of experiencing at least one adverse event was associated with: age (OR: 1.019/year; 95 %CI: 1.00–1.03; p-value = 0.003), female gender (OR: 1.52; 95 %CI: 0.99–2.31; p-value = 0.05), indication for allo-HSCT (OR: 1.67; 95 %CI: 1.004–2.78; p-value = 0.048) and living in a large city or state capital versus smaller cities (OR: 1.92; 95 %CI: 1.26–2.92; p-value = 0.002). Furthermore, the median and mean cumulative adverse event rates were 1 and 1.58 (range: 1–6 per patient). This analysis identified that indication for allo-HSCT (IRR: 1.23; 95 %CI: 1.03–1.49; p-value = 0.03) and age (IRR: 1.004/year; 95 %CI: 1.000–1.007; p-value = 0.04) were variables associated with increased relative incidence ratios of cumulative adverse events. Table 3 summarizes most relevant differences between patients indicated for allo-HSCT and those without indication.