Unmet clinical needs and challenges in the management of acute leukemia in Turkey: a modified Delphi study

Acute myeloid and lymphoblastic leukemia (AML and ALL) are among the hematological malignancies that cause significant morbidity and mortality worldwide. AML is the most common acute leukemia (AL) in adults, constituting about 80% of the cases, with an incidence of about 3–5 cases per 100,000 population [1,2]. ALL also has similar incidence rates of 1–5/100,000, in which about two-thirds of the patients have B-cell phenotype [1,3].

The management of AL has several challenges that create a predisposition toward the unmet needs of patients. Although, in recent years there have been advances in treating these patients and novel targeted and immune therapies have been introduced into the hematological malignancies platform, the older patients still face delayed diagnosis, poor response to therapy, increased recurrence and several co-morbidities that are already present or developed during the treatment [4]. Apart from these clinical challenges, patients may also suffer from paramedical concerns, including psychosocial aspects, communication with physicians, awareness and knowledge of the medical staff about the specific characteristics of these patients, etc. [5]. Medical societies and associations strongly implicate the importance of delivering concurrent palliative care in hematological malignancies [6]. When all these potential challenges are considered, the concept of unmet needs in the management of patients with AL becomes more evident [7]. There is a paucity of studies addressing the unmet needs of patients with AL, and the available ones are focused on specific diagnoses, symptoms or treatment approaches [5,8–11]. When the available data are incomplete or subject to uncertainty, and large-scale observational studies or randomized controlled studies are not able to be carried out due to economic, ethical or pragmatic reasons, researchers use the Delphi technique to generate and aggregate different ideas, achieve accurate predictions, identify expert views or identify a consensus on a specific issue. The technique is based on an iterative process wherein a group of experts is asked to answer a list of questions. Afterward, each expert receives anonymous group results and is given the option to change their previous answers [12].

In this study, we conducted a modified Delphi technique to evaluate the critical points in the management, diagnosis, treatment and follow-up of AL in Turkey and the patient–physician communication in this cohort.

Materials & methods

This study aimed to identify the unmet needs and the barriers in the management of patients with AL in Turkey. For this aim, a modified Delphi technique was utilized to determine the expert consensus on the general questions, patient–physician communication, diagnosis, treatment and follow-up of AL. This modified Delphi approach included the circulation of predetermined questions twice among the participants and the third component, a face-to-face meeting to maximize the expert interaction [13].

The experts’ panel included 13 participants from different private and public academic referral institutions in Turkey. The participants were medical doctors with academical degree and minimum 10 years of experience in the management of AL. After establishing the experts’ panel, a questionnaire was formed based on the most current knowledge in the management of AML (excluding acute promyelocytic leukemia [APL]) and B-cell acute lymphoblastic leukemia (B-ALL). Since 75% of ALL cases are B-ALL [14] and approximately 85% of AML cases are non-APL AML [15], the questionnaire was created for B-ALL and AML (excluding APL). We aimed at getting more specific results by excluding APL since the diagnosis and treatment processes is different for this disease. The consensus thresholds defined by the experts were as follows:

• Complete consensus: 100% agreement (13 of 13 participants);

• Strong consensus: more than 80% agreement (at least 11 of 13 participants);

• Consensus: nearly 70% agreement (at least nine of 13 participants);

• Majority opinion: more than 51% agreement (at least seven of 13 participants).

The number of questions evaluated by the experts in each section was seven in general management, five in patient–physician communication, 20 in diagnosis, 28 in treatment and six in the follow-up sections. The questions and the levels of agreement are presented in the relevant tables (Supplementary Appendix). Five of the 66 questions were open-ended and aimed to obtain answers that were not included in the context of the remaining 61 questions in the AL management journey survey.

In the first half of the online meeting, the participants received a brief introduction on the method of face-to-face discussions, particularly about achieving ‘consensus’ and the interpretation of disagreement topics. Following the briefings, the experts’ panel heard the issues with a complete (100%) or strong (>80%) ‘consensus’ from the two sessions of previous circulations of questions. Then, the participants shared feedback and expressed their thoughts on these issues for a further in-person discussion with the other experts. At the second half, 26 items without ‘consensus’ from previous circulations were reevaluated by the panel.

Results

At the initial round of questions, the participants answered 66 items that were prepared with the support of two hematologists who did not respond the survey, and a consensus was obtained on 25% (15/61) of the issues. The five open-ended questions included responses that overlapped with the remaining 61 questions, and were not redebated in the subsequent rounds. At the second round, participants re-evaluated 46 questions, and a consensus was obtained on 33% (20/61) of the issues. The anonymity was protected in the first two rounds. During the face-to-face discussions, consensus or stronger agreement was obtained on 34% (21/61) of these issues. No consensus was reached on 8% (5/61) of the issues.

For the six questions on the timeframes from the first symptom to diagnosis, to obtain test results and to initiate therapy, the two consecutive time options in the answers were combined and considered as the consensus statement.

General management

For the seven questions in this section, the experts had a ‘majority opinion’ on two questions, a ‘consensus’ on four questions and a ‘strong consensus’ on one question (Supplementary Appendix Table 1).

The ‘majority opinion’ was on the issues that need improvement in the management and patient access to the necessary treatment in AL. Accordingly, the experts panel reported that the feasibility and the duration of the test procedures need improvement, and patient access to new treatments in AL is inadequate.

‘Consensus’ and ‘strong consensus’ were obtained on the current treatment modalities, the knowledge level of the health professionals, education planning for the healthcare staff and the awareness about allogeneic stem-cell transplantation (ALLO-SCT). The experts disagreed that adult AL patients can reach the most current treatment options, healthcare professionals taking a role in treating adult AL together with hematologists have adequate levels of knowledge, and the level of awareness about ALLO-SCT in the population is adequate. The experts agreed that educational activities should be planned every 4–6 months for other healthcare staff and hematologists.

Patient–physician communication

For the five questions in this section, the experts had a ‘majority opinion’ on one question, a ‘consensus’on three questions and a disagreement on one question (Supplementary Appendix Table 2).

The ‘majority opinion’ was about the delivery of adequate information about the treatment to AL patients. ‘Consensus’ was obtained on the statements that the physicians experience difficulty in informing the patients about the diagnosis, have enough time to tell the diagnosis, and inform patients adequately before their treatment is applied. They disagreed on the proportion of patients who are adequately informed about their diagnoses.

Diagnosis

For the 20 questions in this section, the experts had a ‘majority opinion’ on one question, a ‘consensus’ on 14 questions, a ‘strong consensus’ on four questions and a complete ‘consensus’ on one question (Supplementary Appendix Table 3).

The ‘majority opinion’ was about the number of ALL patients treated at a hematology unit per year, which was reported as 11–15. ‘Consensus’ was obtained on the statements about the duration from the first symptoms to diagnosis in AL, the access to tests, and the duration to obtain test results. The ‘strong consensus’ statements were about access to and duration of diagnostic tests, and the only complete ‘consensus’ was about the access to next-generation sequencing (NGS) tests for appropriate AML classification.

The experts agreed that adult AL patients are generally diagnosed in about 3–4 weeks without delays, but about 16–20% of cases may experience delayed genetics-based diagnosis. They also reported that the duration from the first symptom to diagnosis should ideally be less than 2 weeks. In addition, the experts panel stated that the access to diagnostic tests for both ALL and AML is easy. However, the access to genetic and molecular tests for prognostic risk classification, including NGS, is complicated and the test results obtained in a relative long duration (4 weeks in AML) also negatively affect the treatment plans. Other issues with a ‘consensus’ were about timely referral of AL patients to the experts’ centers. The experts agreed that almost all patients can receive their treatment in the hematology centers where they are initially diagnosed; however, about 26–50% of cases were referred to another center mostly due to insufficient hospital bed capacity. Annually, 21–30 adult AML cases is treated at a referral hematology center.

Treatment

For the 28 questions (including four open-ended questions) in this section, the experts had a ‘consensus’ on 21 questions and a ‘strong consensus’ on three questions (Supplementary Appendix Table 4).

Accordingly, the experts stated that the treatment of adult AL patients starts in about less than 3 days following the diagnosis, which is appropriate with the ideal duration of 4–5 days. Nevertheless, not all planned patients, but only about 51–70% of them can undergo ALLO-SCT. With these numbers, the demand is met mostly by the volunteer hematopoietic stem cell (HSC) donors in Turkey. The duration from application to Turkey Stem Cell Coordination Center (TURKOK) for ALLO-SCT to transplantation is about 2–3 months, and 4 weeks of this period is the duration between work-up and transplantation. The primary focus area to increase the donor numbers for improving the ALLO-SCT related issues is reported as increasing awareness in the population. Also, another critical issue is the insufficient number of hospital beds for meeting the demand for hospitalization during adult AL treatment. The estimated average duration of hospitalization for bridging treatments before ALLO-SCT in Relapse/Refractory (R/R) B-ALL is about 3–4 weeks. Experts reported for non fit AML patients that only 10–30% responded to the treatments available in Turkey, and about 71–90% need further treatment modalities other than those available in our country. For the fit AML patients, the response rates are about 51–70%, and the need for further treatments is about 31–50%. The panel also agreed that patients do not receive adequate psycho-oncological support during treatment. The proportion of AML patients who are not suitable for induction treatment is 31–40%. Moreover, only 10–30% of adult AL patients achieve minimal residual disease negativity before ALLO-SCT, and MRD negativity criteria was reported not to be routinely evaluated prior to ALLO-SCT.

Follow-up

For the six questions (including one open-ended question) in this section, experts had a ‘consensus’ on four questions and a ‘strong consensus’ on one question (Supplementary Appendix Table 5).

The experts’ panel agreed that adult AL patients are regularly followed up, and 51–70% of ALL cases and 31–50% of AML cases relapsed during the follow-ups. The proportion of R/R B-ALL and R/R AML patients receiving an ALLO-SCT is about 31–50%.

Discussion

ALs are rapidly progressive diseases that pose a significant disease burden due to the symptoms, treatment related factors, psychosocial deteriorations and other factors associated with the patients’ quality of life (QoL) [16,17]. Therefore, identifying the unmet needs and the challenges in disease management may direct efforts to improve the overall well-being of patients with AL [18]. This study was conducted in this direction and based on the expert opinions gathered using a modified Delphi approach. Circulation of a questionnaire twice and a subsequent face-to-face interview session among the experts to seek the consensus on the issues related to overall management, diagnosis, treatment, follow-up and patient–physician communication provided refined implications regarding the unmet needs that challenge AL management.

In the first half of face-to-face interviews, experts evaluated the issues, which a consensus was obtained in previous circulations, and in the second half, issues without a consensus were discussed to get an agreement if possible. The debates were mainly focused on how questions should be revised to prevent any misunderstanding and the critical points for interpreting the answers correctly. An example for clarifying the questions was regarding TURKOK, application for suitable donor and time to transplantation. Since its foundation, TURKOK’s rising donor capacity facilitated many unrelated donor HSC transplantations, and time interval between donor search and transplant for TURKOK versus other foreign donor registries was significantly shorter. In 2018, according to Arat et al. [19], 55% of the donations were facilitated by TURKOK. The exact definition of work-up and confirmation stages was discussed, and a statement was added in bracket to the question to clarify that the confirmation phase is included in order to interpret the question and answer accurately. Other examples were the questions asking the duration from the first symptom to diagnosis in AL. The ideal timeframe is 2 weeks; however, the experts underlined that the answers to these questions might vary based on the setting or scenario. For example, the subtype of leukemia is critical, and 4 weeks for diagnosis in a patient with CNS involvement and neurological deficit is unacceptable. Second, the physician at the initial treatment site is critical for the delays. It is crucial to know whether the patient is directly referred to a hematologist or a family practitioner, as it is the actual situation in Turkey. They also stated that the questions were prepared irrelevant to the specialty, the consensus answers are appropriate if a family practitioner first sees the AL patient, then an internal medicine specialist and then a hematologist. Another discussion was on the question “What percentage of AML patients are not suitable for standard induction therapy for reasons such as age, comorbidity and performance index?” . Despite the consensus on 31–40% of the statements, panel members implied that this proportion is reflecting a general estimation and certain circumstances should be interpreted carefully. Finally, a discussion was made on two consensus items that opposed with each other. In one of those items, the experts agreed that the knowledge and awareness level in the population about ALLO-SCT is inadequate. However, another item revealed that the number of volunteer HSC donors meets the demand for ALLO-SCT. The experts stated that if the number of volunteers meets the demand, this also means that there is an adequate level of knowledge and awareness, and these consensus items may be regarded as mutually exclusive. Thus, the readers should remember that ‘adequacy’ is a subjective term, and one may disagree with the population’s level of knowledge and awareness. The experts reached a consensus on the statement that volunteer donors meet 51–70% of the need, which may suggest an adequacy, but also a ground to be improved. As a consequence, the experts stated that increasing the awareness in the population to increase the volunteer donor numbers is the most critical issue to meet the need and improve the access to treatment for patients who need transplantation. The first session of interviews finalized with a general recommendation to readers that interpreting the results considering the circumstances in the COVID-19 pandemic should not be a proper approach.

As an overall interpretation of the results from the Delphi rounds, the major challenges in the management of AL determined by the Turkish experts’ panel can be summarized as the limited access to novel treatment modalities and genetic tests, including NGS, for disease and risk classification, limited bed capacity for hospitalized patients, insufficient level of knowledge among allied health personnel working together with hematologists, limited availability of psycho-oncological support and low levels of awareness in the population about the importance of volunteer stem cell donation and ALLO-SCT. The experts also provided varying numbers for the numbers of patients that are being treated in hematology centers; however, this was associated with the country profile, which the numbers vary according to the capacity of different centers (public vs private) in Turkey. Additionally, discussions also revealed that there was adequate time to communicate with the patients and the families about the diagnosis; however, the time was limited to provide detailed information about the course and further possible scenarios. In a similar report from Spain, again using Delphi methodology on AML patients’ epidemiology and management, the authors underlined the need for newer treatments, inclusion of health related QoL to the studies, promotion of information exchange among physicians, centralized complex diagnostic tests, need for cost–effectiveness studies and training of all AML related healthcare professionals by pathology experts [20]. Training of the healthcare workers and centralized diagnostics are valued concordantly in our analysis. Gidman et al. [21] published clinicians’ perspectives on cure in ALL patients with minimal residual disease (MRD). This oligo national study, including the experts from Germany, UK and France used Delphi methodology and defined cure in MRD positive patients according relapse-free survival and longitudinal MRD status. They also emphasized the importance of MRD negative relapse-free survival in ALL patients like our final results. A recent study abstract reported by a Turkish center retrospectively reported median time from diagnosis of AML (excluding APL) to obtain information on PCR results for both NPM-1 and FLT3 mutational status as 8 days [22].

For the limited access to the novel therapies, the experts underlined that compassionate use programs/early-access programs are working well; however, the social security and the reimbursement systems do not cover these treatments yet. Current evidence in the literature also supports this barrier in the world. A recent study evaluating the access to therapy for AML in developing countries reported that the economic burden of treatment is considerably high, and it is a major challenge faced by countries even they have diverse healthcare structures [23]. For the limited availability of the genetic tests, the experts stated that only a few centers could make genetic testing for the appropriate classification of AL according to the WHO criteria and the European Leukemia Network prognostic risk classification criteria [24,25]; however, this is not sufficient when the need and demand are considered.

For the low levels of knowledge among allied health personnel, the experts agreed that the major obstacle is the inadequate capacity of nurses and other personnel to meet the demand in healthcare facilities, which results in frequent circulation of these personnel in different departments. In addition, the rapid shifts do not allow a nurse or others to gain adequate knowledge specific to AL patients. Thus, the experts recommended regular educational and training activities for allied health personnel every 4–6 months. Several previous studies also stated the importance of continuing education for hematology nurses. A study by Liptrott et al. [26] evaluated the educational priorities of hematology nurses about nursing care. They reported that coping with end of life is also a topic of interest in educations for all, but the psychosocial issues are among the education priorities of more experienced staff. This is also consistent with our findings of the inadequate delivery of psycho-oncological support to patients, which might be associated with both the experience and the educational need of the allied health personnel working together with hematologists.

Limitations

This study was susceptible to the overall limitations of a modified Delphi study, mainly due to the fact that the technique has no methodological standardization. Two external hematologists prepared the questionnaire rather than an initial round with the panelists. The low number of experts from referral institutions is a limiting factor in reliability and representation. Having only hematologists as experts and not including other disciplines are other limitations of our study that can influence the robustness and validity of our findings. The results of our study give some ideas for improving the management of AL based on answers of a selected experts in Turkey and cannot be generalized for other countries.

Conclusion

Healthcare system is evolving to a more patient–centric approach and AL patients are not an exception. The challenges in the management of AL from the perspective of healthcare professionals are critical to guide the efforts to enhance the quality of healthcare delivery, as well as improving the QoL of patients. Uncovering the barriers and unmet needs in the management of patients with AL can only be achieved by paying attention on the journey of patients. This Delphi analysis, which is of its first kind for Turkey underlined the importance of communication with the patients to give more satisfactory information in each step of the disease management, regular education programs for healthcare staff, further initiatives to increase access to novel therapies and facilitation of NGS based genetic testing and the need to improve healthcare facilities’ in-patient bed capacity. Further studies to evaluate these challenges from the patients’ and the healthcare providers’ aspects may contribute much to reveal the barriers in AL management in Turkey.