Breast cancer is the most common cancer among women in Slovenia. In 2019, the age-standardized incidence rate was 108.4 per 100,000 women according to the European standard population (1). Over time, the net survival of breast cancer patients in Slovenia has improved significantly, from 77.5% (95% CI 76–79) during 1997–2001 to 87.6% (95% CI 86.3–88.9) during 2011–2015 (2). While survival rates have improved, maintaining quality of life (QoL) during cancer treatment is increasingly prioritized in clinical trials and real-world practice. Enhancing overall survival (OS) or QoL is a key endpoint for the regulatory approval of new therapies (3).

The evaluation of QoL covers different aspects of an individual’s daily life, including physical and emotional well-being, psychosocial functioning and treatment-related toxicity. Information about a patient’s health that comes directly from the patient without the intervention or interpretation of a healthcare professional are patient-reported outcomes (PROs) (4). PROs are assessed via patient-reported outcome measures (PROMs), usually with the Patient-Reported Outcomes version of the Common Terminology Criteria for Adverse Events (PRO-CTCAE) (5) or validated questionnaires. The European Organization for Research and Treatment of Cancer (EORTC) has developed three primary types of questionnaires to evaluate patient well-being: core questionnaires, module questionnaires, and standalone questionnaires. Core questionnaires serve as a basis for assessing quality of life in all cancer patients, whereas module questionnaires aim to increase sensitivity and specificity in specific types of cancer (6). The most commonly utilized PROMs in breast cancer include the EORTC Quality of Life Core Questionnaire 30 (EORTC QLQ C-30) and EORTC Breast Module 23 (EORTC QLQ BR-23) (7, 8). The impact of using PROMs to improve QoL has been demonstrated in several clinical trials, which revealed that early detection of adverse events not only increased QoL but also reduced the hospitalization rate and even prolonged OS (9–14). However, the main limitation of most existing methods of data collection is the recording of data on predefined dates (e.g., at clinic visits, every 12 weeks), which leads to information gaps between assessments and the use of limited sets of preselected symptoms that cannot capture all that actually occurred during therapy. Advances in informatics and mobile technology have enabled electronic patient-reported outcomes (ePROs) in real time via mobile applications (m-apps), offering a potential solution to overcome these limitations. Because the use of PROs even prolongs OS, ePROs are a way to become digital therapeutics (15). The use of the m-app with the PRO-CTCAE as a tool for self-reporting significantly empowers patients to adequately report adverse events of cancer chemotherapy and increases their self-management of treatment-related side effects (16–18). Despite their technological feasibility, the widespread adoption of such m-apps remains limited, probably due to the challenges in managing severe symptoms and physicians’ reluctance to accept new technologies (19).

To investigate the feasibility and utility of ePRO collection with our custom-made m-app OnkoVed (20), we conducted a prospective cohort study in patients receiving chemotherapy. Our primary objective was to assess the frequency and severity of symptoms reported by patients on a daily basis. We also assessed QoL via validated QLQ questionnaires within the m-app. In addition, we investigated the concordance between patient-reported and doctor-reported symptoms. Finally, we evaluated the usability and satisfaction of the m-app by patients and doctors.

This prospective cohort study focused on daily symptom reporting in patients undergoing chemotherapy for early breast cancer. Patients using ePROs reported a significantly higher number of symptoms, and these symptoms were more severe than those reported in the subsequent assessment and interpretation by doctors during the same patients’ next outpatient visit. The symptoms most underreported by doctors were insomnia and dry mouth. We also used two validated questionnaires, the EORTC QLQ C30 and BR23, and compared periodic and daily symptom reporting. Patient and doctor satisfaction with and usability of the m-app were evaluated by patients and doctors. Both expressed high levels of satisfaction, although doctors expressed a need for a plan to manage severe symptoms when they were reported and a concern that they would be able to take over the additional workload to react to severe symptoms.

In the last decade, ePROs have become a hot topic due to advancements in digital technology and the widespread availability of smartphones, making these services globally accessible. The logical next step is the use of ePROs as a communication tool between patients and doctors, particularly in medical treatments such as cancer treatment, where prompt action at the onset of symptoms is crucial. The goal is to ensure the maintenance of patients’ QoL and adherence to their treatment regimens. Several small studies have been performed worldwide; however, the routine use of ePROMs and incorporation of these data into EHRs were still very rare until recently (11, 24, 25). The experience of The Christie NHS Foundation Trust, launched in 2019, stands out as a key example of large-scale ePROM integration into routine oncology care (26, 27). Real-time data entry into the EHR—without relying on separate platforms—enhanced the clinical usefulness for treating clinicians. Reported benefits include fewer hospital admissions, better quality of life, potential survival benefits, and more efficient consultations (26, 28). Later evaluations identified barriers among older patients with limited access to technology and those receiving radical treatment. Clinicians have also requested clearer data displays within EHR (27). Another UK oncology center that uses ePROMs for patients receiving targeted and immunotherapies has shown a reduced need for face-to-face visits (29). A program using the PRO-CTCAE and pain scales during palliative radiotherapy offered automated self-care guidance and guidance for contacting healthcare providers (30). Studies with ePROM implementation highlight benefits such as improved symptom control, patient-provider communication and quality of life (30–32). The European Society for Medical Oncology (ESMO) now recommends PROM-based monitoring after cancer treatment, including near end-of-life care (33).

In this trial, we prospectively assessed daily symptom reports. Over 4–6 months of chemotherapy, patients reported 75 symptoms (median 25 per patient): 67% mild, 30% moderate, and 3% severe, with fatigue, insomnia, stomatitis, myalgias, arthralgias, and diarrhea among the severe symptoms. Our findings are consistent with an Italian study that also identified fatigue as a highly prevalent but often overlooked severe side effect of adjuvant systemic therapy (12). We then compared patient-reported symptoms with doctor assessments at regular check-ups. Doctors reported fewer symptoms (49 vs. 75, median 7 per patient, IQR 6–10) and generally graded them lower ( Figure 3 ). However, Grade 3 symptoms accounted for 3% of the symptoms in both reports. Patients most frequently reported fatigue, insomnia, and dry mouth ( Figure 4 ), with insomnia and dry mouth being symptoms the most underreported by doctors. Our findings align with those of previous studies (9, 34) showing that doctors tend to underestimate symptoms, particularly those where specific treatments are unavailable. Notably, Grade 3 muscle/joint pain, fatigue, and nausea were underestimated (35). Overall, our results suggest that patients report numerous adverse events affecting their QoL. In contrast, doctors prioritize reporting those that influence treatment decisions, such as not prescribing (postponing) therapy, dose adjustments, or supportive therapy adjustments. We acknowledge the potential for assessment bias in the doctor’s grading of symptom, given the retrospective nature of the assessment. The risk was to some extent alleviated by the use of single assessor. While it may enhance consistency, it does not eliminate bias and could introduce a single-assessor bias, where that an individual’s interpretation systematically influences the data.

In addition, our study highlights the effectiveness of the m-app in managing adverse symptoms. The low percentage of Grade 3 symptoms is likely due to the supportive measures provided by the m-app toolkit, which helps patients self-manage mild to moderate symptoms, preventing escalation to Grade 3. This aligns with previous studies showing that PROs enable early recognition and management of adverse events, reducing symptom duration (12, 36). Notably, our patients were symptom-free for 42% of the treatment duration, whereas Antonuzzo et al. (12) reported a symptom-free period of only 15%, likely due to differences in patient characteristics, chemotherapy regimens, or study methodologies. In our previous prospective cohort study, we demonstrated that early breast cancer patients using the m-app toolkit experienced improved quality of life (QoL) both in the first week after chemotherapy and at the end of treatment compared with the control group without application support (21). Similarly, Fjell et al. reported that an interactive application reduced symptom burden and improved emotional well-being during neoadjuvant chemotherapy for patients with breast cancer (37).

There is a worldwide debate regarding the optimal frequency of ePRO assessment during cancer therapy: once-weekly or daily assessment. Both approaches have advantages and limitations, with more precise symptom reporting in daily monitoring and less burdensome reporting weekly. Symptom monitoring empowers patients to actively contribute to problem-solving measures. On the other hand, experiences revealed survey fatigue in patients, which means that they responded less regularly when they became familiar with symptom management. Therefore, strategies on the frequency of symptom recording should be adjusted to a particular setting (active chemotherapy treatment, maintenance treatment, follow-up on chronic therapy) (38). For example, during adjuvant endocrine therapy, 73.2% of patients completed the baseline survey, whereas 69.6% participated in at least one 3-monthly ePRO follow-up survey via a smartphone within the first six months (39). Like our study, Daly et al. utilized daily symptom reporting in adults receiving anticancer therapy. They analyzed the impact of remote monitoring of red alerts (severe symptoms) and reported that 8.7% of patients who triggered a red alert required an acute care visit within seven days, whereas 2.9% of those without a red alert required an acute care visit. However, after six months, only a quarter of the patients continued daily reporting. The most common triggers for red alerts are pain, dyspnea, and functional decline (40).

A key consideration is whether real-time symptom reporting, despite its complexity for both patients and healthcare professionals, has advantages over cross-sectional validated questionnaires such as the QLQ-C30 and BR23. In our study, real-time reporting identified mouth ulcers as the 4th most prevalent symptom ( Figure 4 ), and questions regarding this particular symptom are not available in either C30 or BR23. However, the updated version of BR23, BR42, now includes questions addressing mouth ulcers under items 57 and 58 (soreness and redness in the mouth). Among the 10 most common symptoms reported by our patients, three (dry mouth, taste changes, and hair loss) are covered in BR23, whereas the remaining six are included in C30. Notably, BR23 does not score symptoms individually but rather as part of a composite “systemic therapy side effects” score. Additionally, the timeframe of assessment differs: C30 evaluates symptoms over the past week, whereas BR23 covers the past four weeks. In our view, daily symptom reporting has advantages at the start of a new treatment or for treatments where several side effects are expected, allowing for immediate intervention, whereas C30 and BR23 provide a more comprehensive overview of baseline status, long-term symptom trends over milestones and monitoring adherence to treatment.

Despite the proven benefits, the widespread integration of e-PROs into routine clinical practice faces numerous challenges, including time consumption, logistical issues regarding severe symptom management and physician reluctance to adopt new technologies. With respect to adoption by doctors, four of our doctors included only two patients in the study, other the six included 4–13 patients. In our study, 56% of the doctors chose not to provide feedback on their evaluation and satisfaction with the application. Among the 44% who did share their opinions (4 doctors in total), the majority (75%) expressed satisfaction and supported its integration into daily practice. However, one doctor (25%) provided a strongly negative opinion, viewing the application as a burden that detracts from the valuable time spent with patients without offering significant additional benefits. There was some additional time spent on patients’ check-ups. The initial patient enrolment took an additional half hour from the physician, and the first follow-up visit was extended by approximately 5–10 minutes for report review and app discussion. Subsequent consultation times were not prolonged, as symptom history taking was potentially made more efficient.

In contrast, patients expressed a very positive opinion on the application. They saw it as a tool that provides a sense of security and helps them manage side effects, which is in line with the observations previously reported in different studies and meta-analyses (11, 41–43). This positive perception is reflected in high initial adherence to adverse effect reporting, which gradually declines as patients become more proficient in managing side effects. Similarly, Handa et al. reported a 25% decrease in adherence to m-app usage from the first chemotherapy cycle to the fourth cycle (35). Lee et al. reported that patients with greater expectations of the application’s usefulness and ease of use were more likely to adopt it, although ease of use did not directly impact compliance (39). We are aware, that usability and satisfaction survey feedback was received from 28 of 46 patients (61%), which represents attrition bias. Owing to survey anonymity and postal distribution, we could not increase the percentage of respondents using reminders, causing attrition bias. The delay in performing the survey (6 months poststudy) may have contributed to nonresponse, possibly introducing selection bias—more satisfied users may have been more likely to respond. Nevertheless, frequent users likely provided the most informed feedback. When considering the usability and satisfaction of m-app users, we need to be aware of the critical limitations of our study in terms of lack of representation from older, digital or health-literacy marginalized populations. On the basis of our sample of young and digitally literate women with breast cancer, it is difficult to draw conclusions about the generalizability of the findings to the general population. Among the physicians, 4 out of 10 responded. Nonrespondents may reflect resistance to use ePROMs, as they had not adopted them in routine practice. This may partly explain also the observed difference in symptom reporting grades.

Considering all the advantages and barriers, we believe that ePROs are promising to become an integral part of routine patient assessment in the near future. Only a quarter of the surveyed practitioners reported capturing PROs in routine clinical practice (44). Many countries have experience with various cancer applications, but only a few enable real-time recording and sharing of ePROs with clinicians (26, 29, 30, 45, 46). Despite the known advantages, the acceptance of ePROs in routine clinical care has been hindered by work processes and technology-based challenges (18). Clinic staff reported frustration and dissatisfaction, citing increased workload without perceived benefits for oncological treatment (11). Further investigation is needed to integrate these applications with EHR, emphasizing the importance of implementation standards, certification and user feedback. The inclusion of stakeholders, health professionals, IT specialists, and patient representatives is crucial in the process (47–50). This multidisciplinary approach could improve patients’ perceived health, outcomes, and hospital resource optimization for ePROs. The benefits of remote control of symptoms include proactive early intervention to reduce symptoms, symptom self-management and improved patient–provider relationships (38).

Our study has several strengths and limitations. One of the major strengths, in our opinion, is daily symptom reporting via the m-app, as only a few studies have enabled real-time symptom monitoring thus far. Additionally, our study empowers patients with appropriate tips for self-management of mild to moderate symptoms. Furthermore, it also offers valuable insights into the different perceptions of adverse effects between patients and doctors, as well as their opinions regarding PRO usage. The main limitations, on the other hand, are the small number of participants, the single-center nature of the study and the requirement for smartphone access, which excluded a significant portion of the elderly population. We acknowledge the temporal mismatch between patients’ daily reporting and retrospective grading of doctors’ symptom assessment as a potential confounder. Next, we did not anticipate creating alerts for Grade 3 symptoms; instead, following our usual practice, we instructed patients to visit the nearest emergency department. There were also some increased costs regarding staff engagement (patient education and preparing reports for doctors). Finally, we used our own questionnaire to evaluate the level of satisfaction and usability of the m-app; however, we later discovered that a validated usability questionnaire, the mHealth App Usability Questionnaire (MAUQ), already exists (51). Therefore, we have validated this questionnaire for the Slovenian language for future studies with the m-app.

Our study demonstrates that real-time symptom tracking via an m-app improves symptom detection, with patients reporting nearly twice as many symptoms as doctors do. While patients found the application highly useful (4.5/5 rating), doctors were more hesitant, perceiving ePROs as an additional burden. Despite some reluctance, mobile-based ePRO follow-up is promising, but requires more systematic, inclusive, and scalable implementation studies to establish feasibility and sustainability in routine practice.