Implementation of oncological clinical trials in primary health care in Cuba: a program generalization study

Abstract

Background

Primary health care (PHC) offers a strategic opportunity to conduct research with immunotherapeutic products for cancer treatment.

Objectives

To evaluate the implementation of a program for conducting oncological clinical trials in PHC settings. The implementation objectives focused on assessing structure, process, and outcome indicators against predefined standards; the intervention objectives compared process and clinical outcomes (overall survival, quality of life, and patient satisfaction) in PHC settings against a historical control from secondary care.

Methods

A longitudinal, prospective, analytical implementation study was conducted in Cuba between January 2016 and August 2023. An open-label, uncontrolled, non-randomized, multicenter, phase IV clinical trial using the CIMAvax-EGF vaccine for patients with advanced lung cancer was used as a case study. The implementation was conducted in three phases: pilot study, replication, and generalization. This paper presents the outcomes of the generalization phase. For evaluation, structure, process, and outcome indicators were assessed against predefined standards and compared with a historical control from secondary health care (SHC). An indicator was considered adequate if it met the standard. The overall dimension (structure, process, or outcome) was considered acceptable if ≥ 70% of its indicators were adequate.

Results

The program was implemented in 119 polyclinics nationwide. For structure, 83% (5/6) of indicators were adequate; for process, 83% (5/6) were adequate; for outcome, 100% (4/4) were adequate. All three dimensions met the predefined standards. Process and outcome indicators were comparable or superior to the SHC historical control. Structure and outcome indicators improved after implementation.

Conclusions

The results support the feasibility and effectiveness of implementing oncological clinical trials in Cuban PHC centers, demonstrating a positive impact on the target population.

1. Introduction

In recent decades, the global burden of cancer has increased substantially, posing a critical challenge to healthcare systems worldwide. According to the World Health Organization (WHO), there were nearly 20 million new cancer diagnoses and approximately 10 million deaths in 2020, with projections indicating a continued rise in the coming decades [[1], [2], [3]]. In Cuba, cancer is the second leading cause of death across all age groups, with 26,289 deaths attributed to malignant tumors in 2021 alone [4]. This epidemiological reality underscores the urgent need to explore innovative strategies to strengthen oncological care and enable a comprehensive, system-wide approach to cancer control [5,6].

Cuba's universal, decentralized public health system is structured around three integrated levels of care, with PHC serving as its foundation. PHC in Cuba is delivered through a network of polyclinics and family physician offices, which serve as the first point of contact for the population. These centers are staffed primarily by general practitioners and family medicine specialists, along with nurses, pharmacists, clinical laboratory specialists, psychologists, social workers, and other allied health professionals. PHC centers do not have full-time specialist oncologists or on-site clinical trial units; cancer diagnosis and specialized treatment are typically referred to secondary and tertiary care hospitals. PHC staffs are trained to provide preventive care, health promotion, chronic disease management, and patient follow-up, but have limited experience in conducting clinical trials, particularly in oncology, where protocols require specialized knowledge of Good Clinical Practices (GCP), adverse event (AE) reporting, and data management. While PHC centers have basic diagnostic infrastructure (clinical laboratory and imaging facilities), they lack the specialized equipment and dedicated research personnel typically available in hospital-based clinical trial units. This gap between PHC capabilities and the requirements of rigorous oncological research constitutes the core implementation challenge that motivated the present study [[7], [8], [9]].

In recent years, the development of novel immunotherapies has transformed the oncology landscape. Unlike conventional chemotherapy, which is often highly toxic, requires intensive monitoring, and is administered in short, intensive cycles, immunotherapies generally have a more favorable toxicity profile and are typically used over prolonged periods to maintain disease control. These characteristics—lower toxicity and chronic use—make immunotherapy particularly suitable for delivery in PHC settings, where patients can be followed longitudinally in their communities, reducing the need for frequent hospital visits and minimizing the burden of travel and hospitalization [5]. For this reason, expanding clinical research into PHC settings presents a strategic opportunity to enhance patient access to immunotherapeutic cancer treatments, promote greater community participation in scientific research, optimize the use of local health resources, and facilitate more personalized and continuous care. [10]. However, PHC settings differ significantly from hospital-based in terms of infrastructure, workflow, and daily priorities. Therefore, implementing oncological clinical trials (CTs) in PHC required a tailored program that addressed these contextual differences.

The scientific rationale for this implementation study is twofold. First, regarding the intervention that was implemented: this was a novel program designed to enable PHC centers to conduct rigorous oncological CTs in community settings, using the CIMAvax-EGF therapeutic vaccine—an immunotherapy for advanced non-small cell lung cancer—as the implementation case study. Traditionally, such trials have been restricted to secondary/tertiary hospital care due to the perceived complexity of trial protocols, the need for specialized infrastructure and the requirement for highly trained personnel. This centralization limits patient access, imposes significant travel burdens, and may reduce participation rates, particularly among vulnerable populations [11]. The program was specifically justified by two key considerations: first, immunotherapies have lower toxicity than conventional chemotherapy and are characterized by chronic use, making them well-suited for community-based delivery [5]; and second, there is a need to democratize access to these therapies by bringing clinical trial conduct closer to patients' communities, without compromising scientific rigor, patient safety, or data quality [12]. The feasibility of conducting CTs outside traditional hospital settings was initially tested in a pilot phase (six polyclinics in Santa Clara municipality) and a replication phase (17 polyclinics in Villa Clara province), which confirmed feasibility and preliminary safety, and informed the subsequent refinement and scaling of the program [9].

The program was expected to achieve its effects through a structured, replicable model that included infrastructure adaptation, standardized training, protocol adaptation, standard operating procedures (SOPs), digital infrastructure, and centralized oversight—enabling PHC sites to meet national and international GCP standards.

Second, regarding the implementation strategy: we adopted a phased scale-up approach comprising three sequential stages—pilot, replication, and generalization—grounded in principles of implementation science and informed by the Consolidated Framework for Implementation Research (CFIR) [13]. The strategy comprised three distinct phases, each with specific objectives and learning goals, reflecting a progressive 'learning by doing' model:

Phase 1 – Pilot Study: Conducted in six polyclinics in Santa Clara municipality. The objective was to test the feasibility, acceptability and preliminary safety of the program in a controlled, small-scale setting, identify unforeseen barriers, refine protocols, training materials and monitoring procedures. This phase allowed for iterative adjustments before wider rollout.

Phase 2 – Replication/Diffusion: Expanded to 17 polyclinics across Villa Clara province. The objective was to test the program's transferability to a broader range of PHC settings within the same region, assess the scalability of training and monitoring systems and gather evidence on implementation fidelity across sites with varying baseline capacities.

Phase 3 – Generalization: Scaled to 119 polyclinics nationwide. Based on the positive results and lessons learned from the previous phases, the program was rolled out nationally, with adaptations to regional differences in infrastructure, resources and workforce, while maintaining core program components (infrastructure adaptation for clinical trials, standardized training, protocol adaptations, SOP implementation, digital infrastructure and centralized oversight).

This phased program was expected to achieve its effects by allowing for progressive organizational learning, iterative refinement of implementation components and gradual building of organizational readiness and staff competence—key determinants of successful scale-up identified in implementation frameworks such as CFIR and the Scale-Up Readiness Assessment [14]. The program was specifically designed to manage the inherent risks of large-scale implementation by containing initial uncertainties within the pilot phase, then systematically building evidence and confidence to justify national generalization. Pilot and replication results, published previously, informed adaptations to training intensity, monitoring frequency and site selection criteria [9].

Based on this rationale, the aims of this study were twofold, distinguishing between implementation and intervention objectives. The implementation objective was to evaluate a novel program designed to enable PHC centers to conduct rigorous oncological clinical trials, by assessing structure, process, and outcome indicators against predefined standards. The intervention objective was to assess whether process and clinical outcomes (overall survival, quality of life, and patient satisfaction) achieved in PHC settings were comparable to those obtained in hospital-based settings, using the CIMAvax-EGF therapeutic vaccine for advanced non-small cell lung cancer as the case study, compared against a historical control group.

2. Methods

2.1. Study design and setting

A longitudinal, prospective, analytical implementation study was conducted across 14 provinces in Cuba. Patient enrolment in the PHC program began in January 2016 and concluded in November 2021, with follow-up continuing until August 30, 2023. This implementation study was conducted alongside an open-label, uncontrolled, non-randomized, multicenter, phase IV clinical trial evaluating the CIMAvax-EGF therapeutic vaccine for advanced non-small cell lung cancer (NSCLC) in PHC settings. This clinical trial served as the case study for evaluating the implementation program. [15]. Trial was registered in the Cuban Public Registry of Clinical Trials (https://rpcec.sld.cu/trials/RPCEC00000205-En). The study was guided by principles of implementation science and is reported in accordance with the Standards for Reporting Implementation Studies (StaRI) statement (Table S1)[13].

2.2. Program description and implementation

The program was designed to enable PHC centers to conduct rigorous oncological CTs by leveraging systemic facilitators and mitigating anticipated barriers (Fig. S1). Its key components encompassed the systematic transfer of clinical trial infrastructure and procedures from secondary to primary care settings (detailed in Table S2):

First, infrastructure development involved initial site evaluation visits to identify polyclinic capacity and needs, allocation of physical spaces (CT consultation, storage for investigational products and clinical laboratory) and coordination of special studies at other institutions. Second, multidisciplinary research teams were established at each site, comprising a lead physician, nurse, pharmacist, psychologist, and clinical laboratory specialist, with protected research time allocated to trial activities. Third, a comprehensive training and certification program included a Diploma Course in Oncology for PHC personnel, alongside courses on GCP, Clinical Trials, Ethics and AE, supplemented by scientific workshops and meetings. Fourth, financial support was planned at the municipal level to ensure sustainability of infrastructure, equipment, reagents and medications. Fifth, trial conduct was facilitated by adapting protocols to PHC conditions, developing mechanisms to improve communication between primary and secondary care research teams, prioritizing endpoints that can be determined in primary care, decentralizing biomarker determination to the provinces and implementing SOPs specific to PHC. Sixth, coordination activities were established between the sponsoring center, hospitals, primary care centers and local health administration leadership, with regular meetings to analyze trial progress. Seventh, ethical oversight was strengthened by creating independent provincial-level ethics committees and training their human resources. Eighth, a comprehensive data management system was developed and implemented, including an Electronic Data Capture (EDC) system at polyclinics, electronic Case Report Forms (eCRFs) and user training and certification. Ninth, monitoring and quality control were ensured through on-site and remote monitoring visits conducted by a Contract Research Organization (CRO) with provincial representation of National Coordinating Center for Clinical Trials (CENCEC) and by the development of validated instruments and indicators to evaluate program implementation.

To monitor the implementation process, monthly face-to-face meetings were held between the sponsoring center and the CENCEC coordination team. These meetings enabled periodic analysis of indicators and timely decision-making to address any identified non-conformities. In addition, findings were disseminated through regional or national face-to-face workshops, which were attended by the research teams from all participating institutions, representatives from CIM, CENCEC, the Comprehensive Cancer Control Section (SICC) of the Ministry of Public Health (MINSAP) and other health decision-makers from the provinces.

2.3. Site selection and ethical oversight

Polyclinics were selected in coordination with the Ministry of Public Health based on: (a) structural readiness (facilities, equipment and cold chain); (b) availability of qualified human resources; (c) geographic distribution to ensure equitable access; and (d) established referral links to secondary care. Of the 179 sites initially evaluated, 167 (93%) met national and international GCP requirements [[16], [17], [18]]. Patient enrollment ultimately took place in 119 polyclinics.

The study received approval from the ethics committees of all participating hospitals and polyclinics. In some provinces, ethical review and approval were provided by centralized ethics committees established ad hoc at the medical universities in each province—a distinctive feature of Cuba's research oversight system that facilitates community-based research. The study was conducted in accordance with the Declaration of Helsinki, and written informed consent was obtained from all patients prior to enrollment.

2.4. Evaluation framework and data sources

2.4.1. Indicators evaluation

Structure, process, and outcome indicators, along with their data collection instruments and standards, were validated a priori via a Delphi expert consensus [19] (detailed results not previously published). For structure and process indicators, data were derived from a content analysis of trial documentation, including investigators' curricula vitae, the Investigator Site File (ISF) and monitoring reports and training records. For outcome indicators, data were obtained from the clinical trial databases used as case studies. Operational definitions are detailed in Table S3.

The evaluation employed two complementary comparative frameworks. The first framework assessed whether the program met its predefined quality standards by comparing the structure, process, and outcome indicators against the established thresholds. This determined the extent to which the program achieved its implementation objectives. The second framework assessed whether process and outcome indicators in PHC settings were comparable to those obtained in hospital-based settings. Structure indicators were evaluated solely against the predefined standards, as comparable data from secondary care were not available.

Given the programmatic nature of this implementation study, a concurrent randomized control group was not feasible. Therefore, process and outcome indicators were compared against a historical control group derived from an open-label, controlled, randomized, multicenter, phase III clinical trial evaluating the CIMAvax-EGF therapeutic vaccine for advanced NSCLC in SHC settings, registered at http://rpcec.sld.cu/trials/RPCEC00000161-En.

For outcomes indicators comparison, only patients from the CIMAvax-EGF treatment arm of the historical trial were included. The treatment regimen and patient inclusion criteria were identical to those used in the PHC program: patients with histologically or cytologically confirmed NSCLC, stage IIIB or IV disease, Eastern Cooperative Oncology Group (ECOG) performance status 0–2, and either stable disease or response to first-line oncology treatment (N = 384). Thus, the only difference between the groups was the level of care in which they were treated (PHC vs. SHC).

Structure indicators were designed to assess the readiness and capacity of PHC centers for clinical trial conduct, including: (a) physical infrastructure (dedicated consultation rooms, storage for investigational products and clinical laboratory); (b) material resources (equipment, cold chain and supplies); and (c) human resources (trained personnel, including lead physicians, nurses, pharmacists, and laboratory specialists, with protected research time).

Process indicators were used to examine whether the clinical trial program was conducted as planned. Specifically, the following process indicators were evaluated: (a) recruitment rate (proportion of eligible patients enrolled); (b) retention rate (proportion of participants who withdrew consent to continue in the trial); (c) protocol deviations (non-compliance with the treatment scheme); (d) safety profile (number of grade 3–4 adverse events per enrolled participant); (e) serious adverse events (number of related serious adverse events reported per enrolled participant); and (f) implementation fidelity (proportion of polyclinics with implemented SOPs).

Outcome indicators were designed to assess the effects and impact of the program on patients' health status, including overall survival, quality of life, and patient satisfaction with care received (Table S3).

For outcomes indicators specific methodology for measurements were needed.

Overall survival (OS): was the primary endpoint among the outcome indicators. OS was calculated from the date of trial inclusion to death or last follow-up. The sample size was determined to test non-inferiority of OS in the PHC setting compared to the historical control group from a prior Phase III trial conducted in SHC. The historical control 12-month survival rate was 45%. Assuming a one-sided alpha of 0.05 and 80% power, a sample of 384 patients was sufficient to detect a non-inferiority margin of 10% (i.e., to rule out that the PHC survival rate was more than 10% worse than the historical control). This non-inferiority margin was selected based on previous CIMAvax-EGF trials and was considered clinically acceptable, as it balances the potential loss of efficacy against the benefits of improved access and reduced patient burden associated with PHC-based delivery.

Quality of Life (QoL): was assessed using the global health status/QoL scale (items 29 & 30) of the European Organization for Research and Treatment of Cancer (EORTC) QLQ-C30 questionnaire [20]. The questionnaires were administered before the implementation of the program and subsequently every 3 months. The absolute change at 12 months was calculated, with an improvement of >5 points considered clinically relevant, based on the established threshold defined by Osoba et al. [[21], [22], [23]].

Satisfaction with Care: was measured using the EORTC IN-PATSAT32 questionnaire [24], excluding the inapplicable item 21. The questionnaires were administered by the psychologist of the investigational team 3 months after program implementation. Since satisfaction with care was not measured in the historical control trial, a cross-sectional comparison was made with a contemporary cohort of hospital-based patients who met the same inclusion criteria and were receiving the same CIMAvax-EGF treatment regimen during the same period.

Specific standards were established a priori for each indicator, defining the target level to be achieved. These standards were developed through a multi-step process: (1) a comprehensive literature review of quality benchmarks used in clinical trials and implementation studies; (2) validation through a Delphi expert consensus process; and (3) empirical confirmation of feasibility based on results obtained during the pilot and replication phases. A dimension (structure, process, or outcome) was deemed acceptable if ≥ 70% of its indicators met the predefined standards. This overall threshold was established following the same rigorous procedure, ensuring that both the indicator-specific standards and the dimensional acceptability criterion were evidence-based and contextually appropriate.

2.4.2. Facilitators and barriers determination

The evaluation of implementation facilitators and barriers was informed by the Consolidated Framework for Implementation Research (CFIR). A mixed-methods approach was used, combining quantitative indicator data with qualitative information derived from the content analysis of trial documentation. Data sources included monitoring reports, site visit logs, the ISF, training records, and minutes from the monthly coordination meetings between the sponsoring center, CENCEC, and participating sites. Additionally, feedback from training sessions and scientific workshops was reviewed to capture perceptions of the implementation process. Recurring themes related to facilitators and barriers were identified through an inductive content analysis of these documents. This approach allowed us to complement the quantitative indicator data with contextual insights into the factors that influenced implementation success.

2.5. Statistical analysis

For each indicator, the proportion meeting the predefined standard was calculated. A dimension (structure, process and outcome) was deemed acceptable if ≥ 70% of its indicators were rated adequate. For outcomes indicators specific statistical considerations were needed.

For OS, the primary hypothesis was that the 12-month overall survival rate in PHC settings would be non-inferior to that observed in the historical SHC control group, with a non-inferiority margin of 10%. Specifically, the null hypothesis (H₀) was that the PHC survival rate is more than 10% worse than the SHC control (i.e., PHC — SHC ≤ −10%), and the alternative hypothesis (H₁) was that the PHC survival rate is not more than 10% worse than the SHC control (i.e., PHC — SHC > −10%).

Survival was estimated using the Kaplan-Meier method, and medians with 95% confidence intervals were reported. Comparisons with the historical control used the log-rank test (significance level p = 0.05). The log-rank test was chosen as the primary method for survival comparison because it is the standard non-parametric test for comparing survival distributions and does not require the proportional hazards assumption. Non-significant p-values were interpreted as indicating no statistically significant difference, not as proof of equivalence. Patients with no follow-up were censored at the date of their last clinical contact.

QoL and Satisfaction: Means and standard deviations were calculated. Comparisons with the control group used independent samples t-tests. Missing values were handled using multiple imputation by chained equations (MICE), assuming data were missing at random (MAR).

Analyses were performed using SPSS v25 and the survminer package in R v4.2.1. The sample size was determined by the total number of patients meeting the inclusion criteria within the PHC network during the study period (N = 384). This reflects the pragmatic nature of the trial, which aims to evaluate the intervention under real-world conditions within the available patient population [25]. Nonetheless, this sample provides 80% power to detect a non-inferiority margin of 10% in the 12-month survival rate compared to the historical control, assuming a one-sided alpha of 0.05. No formal sample size calculation was performed for the implementation outcomes, as these were evaluated descriptively across all participating sites, consistent with the exploratory nature of those analyses.

3. Results

3.1. Structure indicators

The implementation program was extended to 119 polyclinics across 113 municipalities in Cuba. These sites enrolled and treated 741 patients with advanced lung cancer. The distribution of enrolled patients varied by municipality, with seven municipalities including 20 or more patients and 57 municipalities including 1-5 patients (Fig. 1).

Fig. 1.

Distribution of enrolled patients by municipality.

Site selection was coordinated with the MINSAP. From 179 polyclinics initially evaluated for structural and human resource readiness, 167 (93%) were approved. Patient enrolment was activated in 119 (72.5%) of the 164 sites formally proposed for the generalization program.

Electronic data capture was successfully implemented in all 167 approved polyclinics, with 668 PHC professionals certified in its use.

Overall, 83% (5 out of 6) of the predefined structure indicators met the target standard, leading to the structure dimension being classified as acceptable (Table 1).

Table 1.

Evaluation of structure indicators in program generalization in primary health care.

Indicator	Result	Standard	Evaluation
Percentage of polyclinics conditioned to conduct clinical trials	93,2% (167/179)	≥90%	Adequate
Percentage of municipalities participating in the clinical trials program	68,9% (113/164)	≥90%	Not adequate
Percentage of completed investigation teams	93,2% (167/179)	≥90%	Adequate
Number of professionals and technicians trained in clinical trials and oncology	93,2% (668/716)	≥90%	Adequate
Percentage of polyclinics with electronic data management	93,2% (167/179)	≥90%	Adequate
Percentage of investigators trained in electronic data management	100% (668/668)	≥90%	Adequate

3.2. Process indicators

For process evaluation, 83% (5 of 6) of indicators met the standard and were rated as adequate (Table 2). Notable findings included improved patient recruitment metrics—a direct benefit of leveraging the extensive PHC network as a facilitator for patient access—and a reduction in protocol deviations compared to the SHC historical control. However, a higher rate of consent withdrawal was recorded in PHC, potentially reflecting the barrier of less experienced sites or the enrolment of a broader, real-world population. A standardized documentary system for clinical trial management was implemented in 100% of participating polyclinics.

Table 2.

Evaluation of process indicators in program generalization in primary health care.

Indicator	Historical Control in SHC	Result in PHC	Standard	Evaluation
. Inclusion rate (patients/month)	5,12 (405/79)	16,1 (741/46)	≥5%	Adequate
Percentage of participants who have withdrawn consent to continue in the trial	10,3% (42/405)	17,7% (131/741)	≤15%	Not adequate
Percentage of protocol deviations (non-compliance with treatment scheme)	6,8% (2/29)	0,8% (6/741)	≤10%	Adequate
Number of grade 3-4 adverse events per number of enrolled participants	3,6% (9/246)	3,7% (27/722)	≤5%	Adequate
Number of serious related adverse events reported per number of enrolled participants	6,0% (2/29)	1.0%* (8/741)	≤5%	Adequate
Percentage of polyclinics with SOPs implemented	0% (0/167)	100% (167/167)	≥90%	Adequate

Leyend: PHC-Primary health care, SHC-secundary health care, SPOs-Standard Operating Procedures.

3.3. Outcome indicators

The impact of the program on patient-centered outcomes was assessed through survival, quality of life, and satisfaction.

The median overall survival for patients in the PHC program met the pre-defined success standard (Table 3). Comparative analysis with the SHC historical control showed no statistically significant difference in median survival times (log-rank test, p = 0.24). Survival rates at specified timepoints did not differ significantly between groups and met the target standards (Fig. S2).

Table 3.

Evaluation of outcomes/impact indicators in program generalization in primary health care.

Indicator	Historical Control in SHC	Result in PHC	Standard	Evaluation
Global survival	10,0 (8,5; 12,0) N = 272	9,6 (8,4; 11,0) N = 384	≥6 months	Adequate
Survival rate at 12 months of follow-up	45% (39%, 51%) N = 272	42% (37%, 47%) N = 384	≥30%	Adequate
Overall Quality of Life	−2,48	9,41	≥ 5 points	Adequate
Mean of patients satisfaction with health services	98,4 ± 5,5 N = 37	94,3 ± 10,5 N = 85	≥80	Adequate

Leyend: PHC-Primary health care, SHC-secundary health care.

For the Quality of Life (QoL) indicator, patients reported a mean increase of 9.41 points on the EORTC QLQ-C30 global health status/QoL scale between baseline and month 12. This exceeds the 5-point threshold for a minimum clinically important difference, indicating a clinically relevant improvement [[21], [22], [23]]. Mean QoL scores at each evaluation timepoint were not statistically different from those in the SHC control (p > 0.05) (Fig. S2).

On the other hand, for Satisfaction with Care scores on the EORTC IN-PATSAT32 questionnaire averaged above 90 points (on a 0-100 scale), which is classified as "very good" [24]. This result exceeded the expert-defined standard and was consistent with satisfaction levels reported in the SHC setting (Table 3).

Overall, 100% (4 of 4) of the outcome indicators met their respective standards and were classified as acceptable.

3.4. Identification of operational facilitators and barriers

The implementation process revealed operational facilitators [26,27], such as the high motivation of PHC teams and the community's positive reception of local trial access, which bolstered recruitment and satisfaction. Concurrently, key barriers were identified, including logistical delays in site activation and a notable discrepancy in the reporting patterns for Serious AEs between PHC and SHC settings, despite similar rates of high-grade toxicity. This suggests a potential barrier related to reporting thresholds or pharmacovigilance processes in the primary care context [16,18].

4. Discussion

4.1. Principal findings in context

This nationwide implementation study demonstrates that decentralizing oncological CTs to PHC is feasible and effective when a structured program actively leverages systemic facilitators and addresses critical barriers. The successful generalization to 119 polyclinics, achieving patient outcomes comparable to a hospital-based historical standard, confirms the viability of this model. The evaluation showed that structure, process, and patient-centered outcome indicators met predefined acceptability thresholds, with a clinically significant improvement in quality of life underscoring a holistic benefit.

4.2. Facilitators of successful implementation

The success of this scale-up was underpinned by several key facilitators:

Contextual Facilitators: Cuba's pre-existing, robust PHC network and its integrated three-tier health system provided the essential infrastructure and formal referral pathways, creating a natural platform for decentralized research [7,8]. The implementation of the program allowed the creation of structural capacities and human resources in the country's polyclinics for conducting oncological CT in accordance with the requirements of GCP [[16], [17], [18]]. These capacities permit the sustainability of the program for the execution of clinical trial protocols with pharmaceutical products or medical equipment used in the treatment or evaluation of oncological patients, which can be used in the context of PHC [[28], [29], [30]].

Programmatic Facilitators: The strategic, phased rollout (pilot → replication → generalization) allowed for iterative learning and protocol refinement. Besides it, the deployment of user-friendly EDC systems was a critical operational facilitator, enabling real-time quality assurance and remote monitoring across a dispersed network [[28], [29], [30], [31], [32]]. Ersek y cols. also highlighted the importance of electronic data management for the development of oncological clinical trials in the community [30].

Human-Centric Facilitators: The "community advantage"—proximity of care to patients' homes—acted as a powerful facilitator for patient satisfaction, quality of life and likely for recruitment and retention, aligning with core principles of patient-centered care [[33], [34], [35], [36]].

4.3. Barriers encountered and mitigated

The implementation process also identified significant barriers, not all fully resolved:

Structural and Knowledge Barriers: The initial lack of clinical trial experience at new PHC sites was a major barrier, mitigated by intensive, ongoing training. The program's dependence on sustained central coordination from CENCEC and CIM, while effective, highlights a potential sustainability barrier for long-term operation.

Methodological Barriers: The use of a historical control, while a pragmatic choice for this implementation study, is a fundamental methodological limitation that restricts causal inference due to potential temporal biases and unmeasured confounding.

Operational Barriers: Discrepancies in the reporting patterns for Serious AEs between PHC and secondary care settings, despite similar rates of high-grade toxicity, point to a barrier in safety surveillance processes or reporting thresholds that requires standardization [37]. Logistical delays in activating all approved sites reflect common scaling challenges.

4.4. Strengths, limitations, and transferability

The primary strength of this work is its unprecedented scale and systematic, phased approach within a real-world trial. We acknowledge that the use of a historical control group is a limitation that may introduce chronological bias due to improvements in supportive care over time. However, the historical cohort was selected from the immediate preceding period using identical clinical protocols and inclusion criteria in secondary care. A key limitation is the focus on a single therapeutic product (CIMAvax-EGF) for lung cancer, which may affect generalizability to other oncologic treatments. Furthermore, the success of this model is partly contextual, facilitated by Cuba's strong, integrated public health network. Direct transferability to other health systems would require careful adaptation of the core components to local infrastructures and capacities.

5. Conclusion

In conclusion, this structured implementation program successfully equipped PHC centers to conduct high-standard oncological CTs, achieving patient outcomes comparable to the hospital-based standard. This model offers a promising pathway to make clinical research more accessible, equitable, and patient-centered.

CRediT authorship contribution statement

Geidy Lorenzo Monteagudo: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Pedro Pablo Guerra Chaviano: Writing – review & editing, Validation, Supervision, Formal analysis, Data curation, Conceptualization. Ramón Alberto Ortiz Carrodeguas: Writing – review & editing, Supervision, Methodology, Investigation, Formal analysis, Conceptualization. Lizet Sánchez Valdés: Writing – review & editing, Validation, Supervision, Methodology, Investigation, Formal analysis, Conceptualization. Abdel Bermúdez del Sol: Writing – review & editing, Supervision, Methodology, Investigation, Formal analysis, Conceptualization. Carmen Viada González: Writing – review & editing, Software, Investigation, Formal analysis. Leisy Nieto Reyes: Writing – review & editing, Supervision, Methodology, Investigation, Conceptualization. Orestes Santos Morales: Writing – review & editing, Supervision, Project administration, Formal analysis. Maria Acelia Marrero Miragalla: Writing – review & editing, Supervision, Methodology, Investigation. Agustín Lage Dávila: Writing – review & editing, Methodology, Investigation, Conceptualization. Tania Crombet Ramos: Writing – review & editing, Visualization, Supervision, Resources, Project administration, Investigation, Conceptualization.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Ethical declarations

Ethics approval and consent to participate in the clinical trial included in this study, were reviewed and approved by ethical committees of hospitals, polyclinics and centralized ethical committees (one per province) created ad hoc for primary care institutions in some provinces, in accordance with the Declaration of Helsinki. Informed consent was obtained from all study participants. Below are the ethical committees that approved the clinical trial:

• 1.“III Congreso” Hospital, Pinar del Río province
• 2.University of Medical Sciences centralized ethical committe of Pinar del Río province
• 3.“Comandante Pinares” Hospital, Artemisa province
• 4.University of Medical Sciences centralized ethical committe of Artemisa province
• 5.“Leopoldito Martínez” Hospital, Mayabeque province
• 6.University of Medical Sciences centralized ethical committe of Mayabeque province
• 7.“Hermanos Ameijeiras” Hospital, Havana province
• 8.“Enrique Cabrera” Hospital, Havana province
• 9.“Benéfico Jurídico” Hospital, Havana province
• 10.“Joaquín Albarrán” Hospital, Havana province
• 11.“Calixto García” Hospital, Havana province
• 12.“Miguel Enríquez” Hospital, Havana province
• 13.“Ángel Arturo Aballí” polyclinic, Havana province
• 14.“Antonio Maceo” polyclinic, Havana province
• 15.“California” polyclinic, Havana province
• 16.“Carlos M. Portuondo” polyclinic, Havana province
• 17.“Efraín Mayor” polyclinic, Havana province
• 18.“Elpidio Berovides” polyclinic, Havana province
• 19.“Federico Capdevila” polyclinic, Havana province
• 20.“Jorge Ruíz Ramírez” polyclinic, Havana province
• 21.“Julio A. Mella” polyclinic, Havana province
• 22.“Lidia y Clodomira” polyclinic, Havana province
• 23.“Luis Galván” polyclinic, Havana province
• 24.“Managua” polyclinics, Havana province
• 25.“Turcios Lima” polyclinic, Havana province
• 26.“Vedado” polyclinic, Havana province
• 27.“Betancourt Ninínger” polyclinic, Havana province
• 28.“José Ramón López Tabranes” Hospital, Matanzas province
• 29.University of Medical Sciences centralized ethical committe of Matanzas province
• 30.“Gustavo Aldereguía Lima” Hospital, Cienfuegos province
• 31.University of Medical Sciences centralized ethical committe of Cienfuegos province
• 32.“Celestino Hernández” Hospital, Villa Clara province
• 33.University of Medical Sciences centralized ethical committe of Villa Clara province
• 34.“Camilo Cienfuegos” Hospital, Sancti Spíritus province
• 35.University of Medical Sciences centralized ethical committe of Sancti Spíritus province
• 36.“Antonio Luaces” Hospital, Ciego de Ávila province
• 37.“Roberto Rodríguez” Hospital, Ciego de Ávila province
• 38.University of Medical Sciences centralized ethical committe of Ciego de Ávila province
• 39.“Manuel Ascunce Domenech” Hospital, Camagüey province
• 40.“María Curié” Hospital, Camagüey province
• 41.University of Medical Sciences centralized ethical committe of Camagüey province
• 42.“Ernesto Guevara de la Serna” Hospital, Las Tunas province
• 43.University of Medical Sciences centralized ethical committe of Las Tunas province
• 44.“Vladimir Ilich Lenin” Hospital, Holguín province
• 45.University of Medical Sciences centralized ethical committe of Holguín province
• 46.“Carlos Manuel de Céspedes” Hospital, Granma province
• 47.“Celia Sánchez Manduley” Hospital, Manzanillo, Granma province
• 48.University of Medical Sciences centralized ethical committe of Manzanillo, Granma province
• 49.“Saturnino Lora” Hospital, Santiago de Cuba province
• 50.“Juan Bruno Zayas” Hospital, Santiago de Cuba province
• 51.University of Medical Sciences centralized ethical committe of Santiago de Cuba province
• 52.“Agostino Neto” Hospital, Guantánamo province
• 53.University of Medical Sciences centralized ethical committe of Guantánamo province

Consent for publication

Not applicable.

Declaration of generative AI and AI-assisted technologies in the writing process

During the preparation of this work, the author used ChatGPT (OpenAI) as a writing assistance tool to improve the clarity, grammar, and flow of the manuscript, and to assist with the translation of certain sections from Spanish to English. After using this tool, the authors reviewed and edited the content as needed and takes full responsibility for the content of the published article.

Funding

This study was funded by the Cuban Ministry of Public Health and the Center of Molecular Immunology.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Data availability

Data will be made available on request.