Effectiveness of Additional Risk Minimization Measures Related to Interstitial Lung Disease/Pneumonitis With Trastuzumab Deruxtecan Treatment of Breast Cancer in European Countries: Results and Learnings From a Post‐Authorization Safety Survey Among Physicians

ABSTRACT

Background

Interstitial lung disease (ILD)/pneumonitis is an important safety risk of trastuzumab deruxtecan (T‐DXd) treatment. This study assessed the effectiveness of additional risk minimization measures (aRMMs) outlined in the educational material (EM) for the product regarding physicians' awareness, knowledge, and implementation related to the risk, early detection, and management of ILD/pneumonitis.

Methods

A web‐based survey was conducted among prescribers of T‐DXd from seven European countries (Austria, Denmark, France, Germany, Sweden, Spain, and the UK) between 03/2021 and 11/2022. The overall effectiveness of the aRMMs was determined as the percentage of physicians correctly responding to questions in the awareness, knowledge, and implementation domains, with cut‐off thresholds of ≥ 80%, ≥ 60%, and ≥ 75%, respectively.

Results

Overall, 172 prescribing physicians from seven countries completed the survey questionnaire (response rate: 3.1%). The majority (73.8%) of the physicians acknowledged receiving EMs. In all, 91.6%, 46.7%, and 76.7% of the physicians correctly responded to questions in the awareness, knowledge, and implementation domains, respectively. The low score observed in the knowledge domain was primarily because the respondents did not acknowledge fever as a typical symptom of ILD/pneumonitis. When a post hoc sensitivity analysis was performed where “fever” was not classified as a mandatory answer to the question about typical symptoms for ILD/pneumonitis, the proportion of physicians achieving the threshold increased to 68.6%.

Conclusion

The aRMMs effectively imparted awareness and supported the management of T‐DXd‐induced ILD/pneumonitis.

Summary.

• A web‐based survey was conducted in seven European countries to assess the effectiveness of additional risk minimization measures (aRMMs) regarding physicians' awareness, knowledge, and implementation for the detection and management of trastuzumab deruxtecan (T‐DXd)‐induced interstitial lung disease (ILD)/pneumonitis.

• Physicians correctly responded to questions in the awareness (~92%), knowledge (~47%), and implementation (~77%) domains. When it was accounted for that 41.9% of physicians did not regard fever as a typical symptom of ILD/pneumonitis in a post hoc sensitivity analysis, the proportion of physicians reaching the knowledge threshold increased to ~69%.

• The aRMMs imparted awareness about the risk of ILD/pneumonitis among patients receiving T‐DXd treatment; they were implemented effectively.

• The physicians had good knowledge of the detection and management of T‐DXd‐induced ILD/pneumonitis.

1. Introduction

Breast cancer is the most frequently diagnosed cancer and the leading cause of cancer‐related deaths among women [1]. In Europe, the estimated number of new breast cancer cases was 557 532 in 2022, leading to 144 439 deaths [2]. Approximately 15%–30% of breast cancers are of the human epidermal growth factor receptor 2 (HER2)‐positive phenotype [3].

Trastuzumab deruxtecan (T‐DXd) is an antibody drug conjugate (ADC) that contains a humanized anti‐HER2 immunoglobulin G1 monoclonal antibody with the same amino‐acid sequence as trastuzumab, covalently linked to deruxtecan (a topoisomerase I inhibitor) [4]. T‐DXd has been jointly developed and marketed by Daiichi Sankyo and AstraZeneca. In a large‐scale Phase 3 clinical trial (DESTINY‐Breast03, n = 524), T‐DXd demonstrated a significant increase in progression‐free survival and doubled the response rate among patients with HER2‐positive metastatic breast cancer compared to another HER2‐directed ADC [5]. T‐DXd monotherapy was approved by the European Medicines Agency (EMA) in January 2021 for the treatment of unresectable or metastatic HER2‐positive breast cancer and later for HER2‐low breast cancers, HER2‐mutant non‐small cell lung cancer, advanced HER2‐positive gastric, and gastroesophageal junction (GEJ) adenocarcinoma [4, 6]. Since 2021, the Marketing Authorization Holders (MAH) of T‐DXd in Europe (i.e., Daiichi Sankyo Europe and AZ) have launched the drug across the continent.

Drug‐related interstitial lung disease (ILD), or pneumonitis, is a well‐known risk associated with T‐DXd, although the underlying biological mechanism is not yet completely understood [7]. ILD refers to a heterogeneous group of diseases characterized by lung injury and pneumonitis, leading to irreversible fibrosis of the lungs [8]. About 15% of patients receiving T‐DXd may be affected by ILD/pneumonitis [9]. Hence, clinicians administering T‐DXd should be well informed on how to monitor and manage ILD/pneumonitis [8].

To minimize the occurrence of severe ILD/pneumonitis in patients receiving T‐DXd, MAH Daiichi Sankyo Europe proposed additional risk minimization measures (aRMMs) for ILD/pneumonitis. In general, aRMMs include interventions (e.g., educational materials [EMs]), control access programs, controlled distribution system, and others) beyond routine risk minimization measures (e.g., Summary of Product Characteristics, SmPC) [4], designed to prevent or reduce adverse reactions [10]. According to the risk management plan for the product—which describes the measures taken to ensure that the benefits of a drug outweigh its risks throughout its use by the intended population—the MAH Daiichi Sankyo Europe developed EMs that include a guide for health care professionals (HCP) and a patient card [7]. The EMs cover information regarding the identification of ILD/pneumonitis, management during therapy with T‐DXd plus crucial points for communication with the patient (including the handover of a patient card informing that this patient is currently treated with T‐DXd). The MAH also proposed to conduct a prescriber survey as a post‐authorization safety study (PASS) to evaluate the effectiveness of aRMMs with the aim to reduce the risk for ILD/pneumonitis with respect to awareness, knowledge, and implementation of the respective EM in seven European countries. The fieldwork for this multi‐country prescriber survey was conducted between March 2022 and November 2022 and was designed to assess physicians' awareness, knowledge, and implementation of aRMMs related to the risk, early detection, diagnosis, and management of ILD/pneumonitis and to measure the effectiveness of the aRMMs across different European countries.

2. Methods

2.1. Study Design

This cross‐sectional multinational survey was conducted among prescribers of T‐DXd from seven European countries (Austria, Denmark, France, Germany, Sweden, Spain, and the UK). This survey used an online multiple‐choice questionnaire to capture physicians' responses to understand the awareness, knowledge, and implementation of aRMMs related to the risk, early detection, diagnosis, and management of ILD/pneumonitis.

2.2. Setting

The survey participants were mainly office and hospital‐based physicians from the participating countries. The survey was performed at least 6 months and not longer than 15 months after the product launch and distribution of the EMs (about 7 months for 1 country, about 12 months for 3 countries and about 15 months for 3 countries). The start date of data collection varied across countries based on the launch date of T‐DXd in the local market (earliest launch date: March 2021) and the date of distribution of the EMs in each country. In Spain, the product launch date was replaced by the date of the Early Access Program. The fieldwork for the survey required about 10–12 weeks in each participating country. The data collection ended in November 2022. The launch dates of T‐DXd and the distribution dates of the EMs in various countries are presented in Table 1.

TABLE 1.

T‐DXd launch schedule in European countries.

Country	Launch date (month/calendar year)	Distribution of EMs
Austria	March 2021	March 2021
France	April 2021	April 2021
UK	April 2021	March 2021
Denmark	June 2021	March 2021
Sweden	July 2021	April 2021
Germany	February 2022	January 2022
Spain a	June 2021	June 2021

Abbreviation: EM, educational material.

^a Point in time refers to the start of the Early Access Program in Spain.

2.3. Survey Participants

The target population for the survey included all physicians who were provided with the EMs for T‐DXd according to an EM distribution list prepared by the pharmaceutical company for each country to cover all HCPs relevant to the product, including physicians of relevant specialties (oncologists or gynecologists). The contact details of physicians on the distribution list were matched with IQVIA's OneKey database to generate the basis for the execution of the recruitment of the respondents for the survey. The physicians were contacted randomly in batches and requested to participate in the survey. The process was continued until the target number of physicians required to participate from each country was reached (up to 60 physicians depending on the country size and the respective physician population) or until all the physicians on the distribution list were contacted. The number of participating physicians was kept between 30 and 60 for each country to ensure proper representation. Physicians were considered for the survey when their contact details technically worked, or when they had not opted out to be contacted, or when they passed the screening questions and had not stopped the survey after the screening questions.

Exclusion criteria for the survey were physicians who were not actively involved in breast cancer treatment. The survey also excluded physicians with a conflict of interest, such as those employed by either pharmaceutical companies or contracted by regulatory bodies like the EMA.

2.4. Web‐Based Survey Questionnaire

All participating physicians received the link to the web‐based questionnaire comprising multiple‐choice and closed questions aligned with the primary and secondary objectives of the study developed based on the EM (available as Supporting Information). The original survey questionnaire was generated in English and tested by three physicians of relevant specialties for its comprehensibility, consistency, and appropriateness of medical terms. The English version of the questionnaire was used in Denmark, Sweden, and the UK. The questionnaire was translated into local languages in Austria, France, Germany, and Spain, and the translated questionnaires were reviewed and approved by local MAH affiliates.

2.5. Study Objectives

The primary objective of this survey was to assess the physicians' awareness, knowledge, and implementation of aRMMs related to the risk, early detection, diagnosis, and management of ILD/pneumonitis during the treatment of breast cancer patients with T‐DXd. The secondary objective of the study was to assess the responses of the physicians to the individual items within the awareness, knowledge, and implementation domains (Table 2).

TABLE 2.

Primary and secondary outcomes and their criteria for overall effectiveness.

Outcomes	Criteria for effectiveness a
Primary outcomes
1. Proportion of physicians being aware of the important identified risk of ILD/pneumonitis	Q3: ≥ 80%
2. Proportion of physicians knowledgeable about the important risk of ILD/pneumonitis	Q4, Q5, and Q7: ≥ 60%
3. Proportion of physicians answering the implementation questions correctly	Q6, Q8, and Q10: ≥ 75%
Secondary outcomes
1. Physicians' awareness of the ILD/pneumonitis risk and its related minimization measures: proportion that selected at least ILD/pneumonitis	Q3: ≥ 80%
2. Awareness of having received the educational material: proportion that selected “yes”	Q2: ≥ 80%
3. To measure physicians' knowledge on the requirement for treatment modifications in case of suspected ILD/pneumonitis: proportion that selected “yes”	Q4: ≥ 80%
4. To measure physicians' knowledge on the requirement to monitor typical signs and symptoms that allow early detection of ILD/pneumonitis (proportion that selected correct responses which are a, b, c, and d)	Q5: ≥ 80%
5. To assess whether physicians implement the recommended talking points to patients at the recommended frequency:
i. Proportion that selected “at each visit” OR “at most visits” OR “occasionally” OR “at first visit”)	Q6.1–5/Q6.8 ≥ 80%
ii. Proportion that selected “At each visit” OR “At most visits”	Q6.6/Q6.7/Q6.9: ≥ 80%
6. To assess the extent to which physicians implement the distribution of the patient card to their patients: proportion that selected “to all patients” OR “to most patients (equal or more than three‐quarters of patients”)”	Q10: ≥ 80%

Abbreviations: EU, European Union; ILD, interstitial lung disease; PAS, Post authorization safety.

^a According to values taken for respective healthcare professionals' surveys published in the EU‐PAS registry (https://catalogues.ema.europa.eu/).

2.6. Statistical Analysis

Descriptive statistical methods were used to analyze the responses received from the physicians. Absolute and relative frequencies were calculated for categorical variables. In cases of questions with multiple options, the frequency of each option provided by the physicians was reported in the statistical results. A weight variable was applied to each statistical unit (i.e., each analyzable physician's questionnaire) to adjust the effect of under‐ or overrepresentation of participants from individual countries. To calculate the weights, both the observed sample distribution (per country) and their distribution from the “distribution lists” were used. These final weights per stratum were normalized so that the sum of the weights across the whole dataset matched the overall sample size of physicians. All statistical units (i.e., physicians/questionnaires) within the same stratum were assigned the same final weight.

The results of the overall effectiveness of aRMMs are presented as weighted percentage, whereas country‐specific results are presented as unweighted percentage. Confidence intervals of 95% were calculated for weighted results. Complete response rate was calculated as a ratio of the number of physicians who completed the survey to the number of physicians in the pool for the survey. Refusal rate was calculated as a ratio of the number of physicians who refused to participate to the number of physicians in the pool for the survey.

The statistical analysis was conducted using Statistical Analysis Software (SAS) Enterprise Guide 8.2 (SAS Institute, North Carolina, USA).

The study protocol and questionnaire were reviewed by the European Medicines Agency.

2.6.1. Assessment of Effectiveness

The effectiveness of aRMMs was analyzed in three domains: awareness, knowledge, and implementation. For assessing effectiveness at the individual level, each participating physician was asked a total of seven questions/question topics (including sub‐questions/different scenarios), of which one question was from the awareness domain and three questions/question topics each were from the knowledge and implementation domains. In addition, three questions referred to the receipt of EM, the usefulness of the HCP guide, and the experience with the product described through the number of patients treated. The individual physician's domain scores were calculated as a percentage of correct answers provided by an individual physician. The cut‐off thresholds for effectiveness at the individual level in each domain were: awareness domain, 100% of correct answers; knowledge domain ≥ 80%; and implementation domain ≥ 80%. These thresholds align with those used in respective PASS prescriber surveys to evaluate the effectiveness of aRMMs [11, 12, 13].

The overall effectiveness of the aRMMs in each domain was calculated based on the proportion of physicians who correctly answered the questions, where the number of physicians who achieved the cut‐off thresholds for effectiveness at the individual level was used. The cut‐off thresholds for overall effectiveness for each domain (≥ 80% for awareness, ≥ 60% for knowledge, ≥ 75% for implementation domains according to values taken for respective HCP surveys published in the EU‐PAS registry) of the primary and secondary outcomes are presented in Table 2.

2.6.2. Post Hoc Sensitivity Analyses

Two post hoc sensitivity analyses of the primary outcome were performed to analyze the impact of categorization of certain answer options as mandatory in the survey. These analyses were performed in two parts: post hoc sensitivity analysis A and post hoc sensitivity analysis B. In the post hoc sensitivity analysis A, fever was not considered a mandatory response to the question on signs and symptoms of ILD/pneumonitis. Post hoc sensitivity analysis B acknowledged that the patient card may be distributed by HCPs other than the prescribing physicians, depending on the respective healthcare systems of the study countries. The impact of these considerations on the study outcomes was examined.

3. Results

3.1. Participants

In total, a pool of 5476 physicians in the seven target countries was the basis for the survey conducted. Out of this pool, 5110 physicians did not respond to the invitation and 149 refused to participate (refusal rate: 2.7%). A total of 217 physicians agreed to participate (participation rate of 4.0%). Overall, 172 T‐DXd prescribing physicians completed the questionnaire (complete response rate: 3.1%) and were included in the analyses (Figure 1). In the study, the majority (89.5%) of the participating physicians were oncologists, and the remaining (10.5%) were gynecologists. All physicians (100%) from Denmark, Sweden, Spain, and the UK were oncologists. In addition, 98.2% of the physicians from France, 77.8% from Austria, and 70.5% from Germany were also oncologists. The gynecologists were from France (1.8%), Austria (22.2%), and Germany (29.5%). The majority of participating physicians worked in the hospital setting (76.7%) and had over 10 years of experience in healthcare practice (79.1%). Most (70.3%) of the participating physicians had treated 1–9 patients with T‐DXd at the time of responding to the questionnaire (Table 3).

FIGURE 1.

Overview of physician recruitment process. Non‐eligible physicians: Physicians who were confirmed to be non‐eligible to participate in the survey after reaching out to them, e.g., failed the screening. Physicians excluded for other reasons: Physicians who were not considered due to other reasons. Physicians who refused to participate: Physicians who explicitly indicated their refusal to participate.

TABLE 3.

Background information of participants.

Background information	Total (n)	Proportion (%) of physicians
Total number of physicians	172
Specialty
Oncology	154	89.5%
Gynecology	18	10.5%
Care setting
Office based	23	13.4%
Hospital‐based	132	76.7%
Both, office, and hospital to equal amounts	15	8.7%
Other	2	1.2%
Experience in practice
≤ 1 year	0	0.0%
1–5 years	6	3.5%
5–10 years	30	17.4%
≥ 10 years	136	79.1%
Number of patients treated with T‐DXd
1–4	58	33.7%
5–9	63	36.6%
10–20	36	20.9%
≥ 20	15	8.7%

Abbreviation: T‐DXd, trastuzumab deruxtecan.

3.2. Overall Effectiveness of aRMMs

The survey findings suggest that while 172 responders across different countries demonstrated high levels of awareness and implementation regarding the aRMMs, the level of knowledge was lower. In this survey, 91.6% of the responders accurately answered questions related to awareness, 46.7% to knowledge, and 76.7% to the implementation domain (Figure 2). Based on country‐specific data, the proportion of responders who correctly answered questions specific to the awareness domain ranged from 85.7% in the Nordic countries to 100.0% in Spain; the knowledge domain ranged from 30.4% in Spain to 52.9% in the UK; and the implementation domain ranged from 50.0% in the Nordic countries to 94.4% in Austria (Figure 2; unweighted percentage).

FIGURE 2.

Results of main analysis: Physicians' awareness, knowledge, and implementation of aRMMs related to the risk, early detection, diagnosis, and management of ILD/pneumonitis. Effective thresholds: awareness ≥ 80%, knowledge ≥ 60%, implementation ≥ 75%. Overall data are presented as weighted percentages. Country‐wise data are presented as unweighted percentage. ILD, interstitial lung disease; UK, United Kingdom.

3.3. Outcomes in Individual Domains

The effectiveness of the aRMMs in each of the individual domains, awareness, knowledge, and implementation, was determined using predefined thresholds.

3.3.1. Awareness

Most physicians were aware of the risk of ILD/pneumonitis related to T‐DXd treatment (91.6% [Table 4]), and 73.8% of the physicians acknowledged that they had received the EMs. In total, 91.6% of the responders accurately answered questions and achieved the predefined effectiveness threshold (≥ 80%) (Figure 2). With respect to the individual countries, four countries (Austria, France, Germany, and Spain) reached levels of > 90.0%, and the remaining were over 85.0% (Denmark, Sweden, and UK).

TABLE 4.

Responses to individual questions of awareness, knowledge, and implementation domains.

Domain	Topics		Weighted % Physicians with correct answers (% CI)
Awareness	Q3	Specific risks of T‐DXd treatment	91.6% (86.6%–96.7%)
Knowledge	Q4	Recommended measures to manage T‐DXd‐induced ILD/pneumonitis	55.7% (47.0%–64.4%)
Knowledge	Q5	Signs and symptoms typically indicative of ILD/pneumonitis a	36.2% (27.5%–44.8%)
Implementation	Q6	Specific tasks related to providing information and guidance to patients and performing routine checkups
		Task 6–1: Inform the patient that they may experience serious and potentially fatal side effects of lung problems	98.8% (96.8%–100.0%)
		Task 6–2: Check if the patient has a history of ILD/pneumonitis, a history of lung comorbidities, or a history of corticosteroids treatment	98.5% (96.4%–100.0%)
		Task 6–3: Inform the patient that early diagnosis and appropriate management of events of ILD/pneumonitis are essential to minimize serious outcomes	98.8% (96.7%–100.0%)
		Task 6–4: Instruct the patient to contact the physician immediately if they experience even mild signs or symptoms (e.g., cough, dyspnea, fever, and/or any new or worsening respiratory symptoms), as some events can worsen rapidly if not treated	100.0%
		Task 6–5: Instruct the patient not to treat their own symptoms	91.6% (86.5%–96.8%)
		Task 6–6: Checking for signs and symptoms of lung problems	93.3% (88.8%–97.8%)
		Task 6–7: Remind the patient that early diagnosis and appropriate management of lung problems are essential to minimize life‐threatening complications	67.0% (58.7%–75.2%)
		Task 6–8: Remind the patient to carry the patient card at all times	88.0% (82.8%–93.2%)
		Task 6–9: Remind the patient of the importance of adhering to scheduled appointments	71.9% (64.1%–79.8%)
Knowledge	Q7	Actions to be taken by the physician under the following scenarios when no other potential causes of ILD/pneumonitis have been identified
		Scenario 1: ILD/pneumonitis of severity grade 1 if resolved in 28 days or less from the date of onset	94.8% (91.1%–98.5%)
		Scenario 2: ILD/pneumonitis of severity grade 1 if resolved in greater than 28 days from the date of onset	75.6% (68.2%–83.1%)
		Scenario 3: ILD/pneumonitis of severity grade 2 or higher	67.1% (58.8%–75.4%)
Implementation	Q8	Immediate actions to be taken by the physician to manage a suspected T‐DXd related ILD/pneumonitis	47.5% (38.7%–56.3%)
Implementation	Q10	Handing out of the filled patient cards to patients receiving T‐DXd	72.5% (64.9%–80.1%)

Abbreviations: CI, confidence interval; ILD, interstitial lung disease; T‐DXd, trastuzumab deruxtecan.

^a Symptoms provided as option: new onset of dry cough, shortness of breath, new breathing problems, fever, weight gain, abdominal pain, taste loss, coughing up blood.

3.3.2. Knowledge

Physicians were assessed on their knowledge regarding the risk of ILD/pneumonitis and the appropriate actions required to manage the condition depending on the grade of ILD/pneumonitis. More than half (55.7%) of the total number of responders correctly reported the recommended measures to manage the risk of T‐DXd‐induced ILD/pneumonitis, and most (67.1%–94.8%) physicians correctly identified the appropriate actions for ILD management (Table 4).

In addition, physicians were asked about typical symptoms of ILD/pneumonitis. Over 90% of the physicians identified “new onset of dry cough,” “shortness of breath,” and “new breathing problems” as typical symptoms of ILD/pneumonitis. However, “fever” was recognized by only 41.3% of the physicians.

Taken together, the proportion of responders who correctly answered the questions in the knowledge domain (46.7%) was below the predefined effectiveness threshold of ≥ 60% (Figure 2). To substantiate that the reason for failing the effectiveness threshold was the answer to “fever” being a typical symptom, a post hoc sensitivity analysis was performed, where “fever” was not classified as a mandatory answer to the question about typical symptoms for ILD/pneumonitis. This led to the result that the vast majority of physicians had answered this knowledge question correctly. The percentage of physicians who achieved the threshold for the knowledge domain increased from 46.7% (in the main analysis) to 68.6% in this post hoc sensitivity analysis setting (Figure 3a). Overall, an effectiveness threshold of ≥ 60% in the knowledge domain was achieved among all countries following the adjustment for fever, except in France (58.9%) (Figure 3a; unweighted percentage).

FIGURE 3.

Effect of sensitivity analysis on the knowledge and implementation domain. (a) knowledge domain, and (b) implementation domain. Overall data are presented as weighted percentage. Country‐wise data are presented as unweighted percentage. “Fever” was not considered as a typical symptom indicating pneumonitis/ILD in sensitivity analysis A. Distribution of patient cards by healthcare professionals other than the prescribing physicians was considered as a valid practice in sensitivity analysis B. ILD, interstitial lung disease; UK, United Kingdom.

3.3.3. Implementation

Physicians' responses to questions on the immediate actions they would take to manage suspected T‐DXd‐related ILD/pneumonitis, conduct regular clinical tests related to ILD/pneumonitis, and provide information and guidance to patients were examined.

Nearly half (47.5%) of the total physicians correctly identified the immediate actions required to manage suspected T‐DXd‐related ILD/pneumonitis. Also, over 90% of the physicians reported that they would perform routine checkups for ILD/pneumonitis and provide related information and guidance to patients receiving T‐DXd therapy (Table 4). Overall, a total of 76.7% of physicians correctly responded to the questions specific to the implementation domain (Figure 2), and the predefined effectiveness threshold of 75% was achieved in this domain.

As the EM included the handling of patient cards, the study investigated the extent to which the physicians implemented the distribution of the patient cards to patients. The majority (72.5%) of the physicians handed the filled patient card either to most or all the patients (Table 4). However, 9.9% of the physicians reported that the patient card was handed out by another HCP (nurse, pharmacist, fellow physician) to the patient, and a small proportion of physicians (9.3%) never handed out the patient card. Post hoc sensitivity analysis B acknowledged that handing out EMs by HCPs other than the prescribing physicians is a valid practice.

Following the post hoc sensitivity analysis, the percentage of all physicians who achieved the threshold for implementation increased from 76.7% (main analysis) to 79.3% (post hoc sensitivity analysis B) (Figure 3b). However, according to the country‐specific results, only two‐thirds (64.3%) of the physicians achieved the threshold in the Nordic countries. This is higher than the outcome in the main analysis (50%) yet below the threshold level of 75%. In Austria, Spain, and the UK, the results did not change, whereas in France and Germany, the percentage increased marginally (Figure 3b; unweighted percentage).

3.4. Physician Opinion of Utility of EMs

Overall, more than 90% of the physicians found the EMs useful, with 44.2% considering them “very useful” and 47.1% “useful,” whereas 1.2% found them “less useful,” and 7.6% did not provide any response.

4. Discussion

This study assessed the effectiveness of the EMs on the risk of T‐DXd‐induced ILD/pneumonitis and its associated aRMMs through a survey among physicians prescribing T‐DXd from seven European countries.

The study found that aRMMs were effective in the awareness domain as most physicians (91.6%) provided correct answers to the domain‐specific questions and achieved the predefined effectiveness threshold of ≥ 80%. The awareness levels were relatively homogeneous across countries (85.7%–100.0%).

Although the participating physicians had good knowledge of the detection and management of T‐DXd‐induced ILD/pneumonitis, the knowledge domain did not meet the predefined threshold of ≥ 60% for effectiveness, as only 46.7% of physicians provided correct answers to questions in this domain. An evaluation of aRMMs related to atezolimab for the treatment of urothelial cancer and non‐small cell lung cancer published by Patel et al. [14] reported that 39.4% of physicians achieved knowledge scores ≥ 70—the predefined threshold used in this study. Although the thresholds are not directly comparable, as our survey used a knowledge threshold of ≥ 80%—the results are within the same range. Iranzo et al. [15] assessed the effectiveness of risk minimization measures for sodium oxybate prescribed for narcolepsy with cataplexy and reported a wide variation in physicians' correct responses to knowledge‐based questions, ranging from 32.3% to 96.8%. This range is comparable to the 36.2% to 94.8% of physicians correctly answering individual knowledge domain questions in our survey.

A detailed examination of our survey results showed that the majority (> 90%) of the physicians identified new onset dry cough, shortness of breath, and new breathing problems as distinctive symptoms of ILD/pneumonitis; however, “fever” was not considered a specific symptom of ILD/pneumonitis by more than 50% of the physicians. This observation matches expert opinions that “fever” is considered a generic symptom across multiple diseases rather than a specific symptom for ILD/pneumonitis [16]. Thus, in the post hoc analysis, “fever” was not considered a mandatory response to the question related to the specific signs and symptoms indicating pneumonitis/ILD, resulting in the effectiveness threshold (68.6%) being achieved in the knowledge domain.

Most physicians (76.7%) provided correct responses to the implementation domain questions, and the predefined effectiveness threshold of ≥ 75% was achieved. This finding aligns with results of the atezolimab survey reported by Patel et al. [14], where 74.8% of physicians achieved the pre‐defined threshold of ≥ 70 regarding behavior on key concepts of the aRMMs.

In most countries, the predefined threshold was met, except in the Nordic countries and France, where 50.0% and 69.6% of the physicians, respectively, provided accurate responses. Such differences in the implementation may arise due to varying healthcare practices across countries. In this study, 9.9% of the physicians reported that the patient card was handed out by another HCP. The distribution of patient cards by HCPs other than the prescribing physicians is considered a valid practice in some European healthcare systems. Published studies indicate that HCPs and nurses in particular may be involved in a wide range of healthcare activities including patient education, monitoring safety and adherence to medications, coordinating care, and handling prescriptions [17]. Nursing in particular is prioritized and is an integral part of the healthcare system across the Nordic countries [18, 19]. In the post hoc sensitivity analysis in our study, the distribution of the patient card by another HCP was considered a “correct” practice (in contrast to the main analysis where it was deemed “neutral”). Following the analysis, the implementation domain score improved remarkably in the Nordic countries, from 50.0% to 64.3%, and slightly in France, from 69.6% to 71.4%.

The present study provides valuable insights into the effectiveness of aRMMs among T‐DXd prescribers across countries in Europe. Most prescribers (~91%) considered that the aRMMs, described in the EMs, are either “useful” or “very useful.” This result aligns with healthcare professionals' evaluations of the prescriber guide (95.6%) and the patient alert card (97.6%) as part of the aRMMs for apixaban—a medication used for the prevention of stroke, systemic embolism, and venous thromboembolic events—where the majority rated these tools as either very or quite useful [20]. It was observed that most of the physicians were aware of the risks of T‐DXd‐induced ILD/pneumonitis and had implemented aRMMs. This is supported by a post‐marketing report that indicated a high awareness and implementation rate among T‐DXd prescribers, and a stable rate of ILD cases was reported in the European Union [21]. The prescribers also had good knowledge of the risks and symptoms of T‐DXd‐induced ILD/pneumonitis. The study highlighted that fever is not considered a typical symptom of ILD/pneumonitis by most physicians. In addition, due to differences in healthcare systems across Europe, patient cards may be distributed by HCPs other than the prescribing physicians in some European healthcare systems. Hence, a thorough knowledge of the diversity of European healthcare systems is crucial for interpreting study results.

The primary limitation of this study was its low response rate (3.1%), which is consistent with the response rates reported by other post‐authorization safety (PAS) surveys among HCPs evaluating the effectiveness of aRMMs [14, 20, 22]. In agreement with our findings, an earlier meta‐analysis reported that 16 out of 23 similar surveys had a response rate ≤ 10%, with a pooled response rate of 2.1% or 4.7% depending on the statistical models used [22]. Other limitations of this study are those inherent to all web‐based surveys, such as participation bias, stakeholder bias, and social desirability bias. Although a few measures were taken (like random sampling from the EM distribution lists and exhaustion of the complete distribution list were necessary to reach the sample size per country) to minimize these limitations, their effect on the study might not be completely ruled out. In addition, the generalizability of the study is limited to the specialty of the targeted physicians (i.e., oncologists and gynecologists in some countries).

5. Conclusion

The results of this study suggest that the aRMMs described in the EMs were effective in the awareness and implementation domains. The participating physicians had good knowledge of the respective risk management measures for the early detection, diagnosis, and treatment of T‐DXd‐induced ILD/pneumonitis. A good understanding of the diversity of the healthcare system in Europe in the context of the study design stage is of high importance for a correct interpretation of the results. Furthermore, the findings support future discussions regarding regular revision initiatives of aRMMs to further improve the management of T‐DXd‐induced ILD/pneumonitis in real‐world settings.

5.1. Plain Language Summary

Trastuzumab deruxtecan (T‐DXd) is a treatment option for metastatic breast cancer in Europe. Interstitial lung disease (ILD)/pneumonitis is a side effect of T‐DXd treatment. An online survey was conducted among physicians from seven European countries (Austria, Denmark, France, Germany, Sweden, Spain, and the UK) prescribing T‐DXd between March 2021 and November 2022. This survey was conducted to assess awareness, knowledge, and implementation of additional risk minimization measures (aRMMs) among physicians for the detection and management of ILD/pneumonitis deemed related to T‐DXd treatment. The study suggests that physicians had good awareness of the risk of ILD/pneumonitis among patients receiving T‐DXd treatment and implemented the aRMMs accordingly.

Author Contributions

A.W.F., U.S., and B.E. participated in the study conception and design. B.E. was involved in data collection and in statistical analysis. E.B., A.W.F. A.G., and B.E. participated in the interpretation of the results. All authors were involved in manuscript draft preparation, reviewed the results, and approved the final version of the manuscript.

Ethics Statement

Ethical approval was not required to ascertain the effectiveness of the aRMMs in the participating countries. The study was conducted in accordance with all relevant regulatory requirements, including the Guidelines for Good Pharmacovigilance Practices.

Conflicts of Interest

A.W.F., U.S., and E.B. are employees of Daiichi Sankyo Europe.

Supporting information

SUPPORTING FILE 1. The Web‐based survey questionnaire.

SUPPORTING FILE 2. Health Care Professional Guide: ENHERTU (trastuzumab deruxtecan)—Version January 2021.

Wientzek‐Fleischmann A., Stellmacher U., Beyerlein E., Gohlke A., and Ehlken B., “Effectiveness of Additional Risk Minimization Measures Related to Interstitial Lung Disease/Pneumonitis With Trastuzumab Deruxtecan Treatment of Breast Cancer in European Countries: Results and Learnings From a Post‐Authorization Safety Survey Among Physicians,” Pharmacoepidemiology and Drug Safety 34, no. 9 (2025): e70206, 10.1002/pds.70206.