The considerations and approaches to standardised toxicity management that have been outlined are intentionally developed to result in a semi-standardised outcome. Standardisation should be prioritised where evidence supports class-wide guidance, and structured flexibility can be maintained for justified, drug- or target-specific deviations. This approach ensures safety-first, conservative principles whilst keeping toxicity management fit for purpose in a rapidly evolving development landscape. The aforementioned framework was primarily designed with AZ trials as a template, with the potential to tailor to external applicability. Similarly, this article and examples described are focussed on the Topo1i ADC therapeutic class, however extension to other drug classes and modalities is possible. For instance, this framework is being applied to the next generation of AZ ADCs in early development with different payload and antibody targets.

Whilst standardised TMGs are important for quality clinical trial design and consistent AE-related responsive care for all participants, there are several factors, such as age, comorbidities, and lifestyle, that should also be considered by the investigator when putting the guidance into practice. Understanding which patient populations are more at risk to certain toxicities, and considering a holistic overview of the individual patient, is crucial for identifying where application of these tailored treatments is necessary. Nonetheless, the standardised guidelines should form the basis of any approach and can be built upon according to these added risks. Due to these factors being highly specific, general recommendations have been avoided in this article in favour of context-specific patient management. Although not mandatory, the additional aspects discussed in the following sections can be incorporated to increase robustness and applicability of resulting guidelines. While not an exhaustive list, examples of considerations are laid out ahead.

It is also important to consider the scope of TMG standardisation beyond the oncology therapy area and how this may differ across indications. Outside of oncology, drugs are often developed to treat conditions that are chronic in nature. Therefore, the safety profile should enable long-term symptom control whilst maintaining an adequate quality of life. This means that both oncology and non-oncology drugs require a tailored consideration of toxicity and benefit–risk impact. For products being investigated in oncology, often more severe toxicities are tolerated due to the disease being highly associated with morbidity and mortality. By contrast, non-oncology drugs are often treating chronic conditions such as chronic obstructive pulmonary disorder (COPD) that prioritise long-term management of the condition and maximising quality of life, which results in greater emphasis on minimising chronic toxicities to ensure long-term tolerability. Phase I non-oncology studies are often conducted in healthy volunteers, highlighting a large contrast in benefit–risk considerations between therapy areas for early-development studies. For example, interruption or discontinuation criteria for a certain toxicity may be more stringent for a COPD patient. More intensive monitoring for hepatotoxicity may be required if the study drug has a known potential of DILI liability. Key risks of drugs such as liver toxicity and more specifically DILI may be approached differently across therapy areas [14]. Eligibility criteria in a lung cancer trial may allow a wider eligibility window for liver function screening due to liver metastases in the patient population. However, as a result, a study in this indication may require more intensive liver function testing throughout the study due to the toxic nature of oncology drugs when compared with a COPD trial.

Also, grading the severity of these toxicities differs dependent on the therapy area, modality, or indication. Oncology widely utilises the quantitative CTCAE grading system, whereas other therapy areas such as vaccines use a more qualitative mild-moderate-severe scale [15]. Also, for next-generation therapies such as CAR-T therapy, new grading systems are being defined for toxicities emerging with these therapies (e.g. ASTCT Consensus grading for cytokine release syndrome [CRS] and Immune Cell Associated Neurotoxicity Syndrome [ICANS]) [16].

The treatment population, and importantly, the stage of the indicated disease, may influence how patients tolerate treatment and, in turn, how these toxicities are managed. In the first instance, patients with advanced-stage cancer may have compromised organ function due to tumour burden, prior lines of anti-cancer treatment, or metastasis, reducing their ability to tolerate additional toxicities. For those with advanced stage cancer, maintaining quality of life becomes a priority over aggressive treatment, meaning more conservative management of toxicities may be necessary to avoid further deterioration.

Conversely, patients with early-stage cancer or receiving first-line therapy, especially those on treatment with a curative intent, might have better organ reserve and overall health, and therefore the ability to tolerate more severe adverse effects without the need for discontinuation of the IMP. Additionally, more aggressive management of toxicities may be justified in order to achieve efficacy of the treatment. That said, patients with cancer of any stage may endure symptoms that significantly impact their daily lives. Therefore, above all else, it is crucial to prioritise the management of a treatment plan that enhances the patients' quality of life and upholds their best interests throughout the trial. The standardised TMG provides a clear foundation of toxicity management according to best clinical practice upon which investigators build a tailored approach depending on the individual patient and indication.

Another aspect to consider is the risk of the same toxicity in different indications. For example, a patient with non-small cell lung cancer (NSCLC) experiencing drug-induced interstitial lung disease (ILD) as an adverse effect, may have severely impacted respiratory function as a baseline, necessitating careful monitoring and management, perhaps even more so than as laid out in the standardised TMG, with a lower threshold for discontinuation. Conversely, a patient with melanoma and ILD, for example, typically won’t have the same baseline respiratory compromise as NSCLC patients, potentially resulting in a less severe impact on quality of life. The treatment approach to this toxicity would still be similar to NSCLC, but the focus might be more on monitoring and treating symptoms rather than discontinuation of the IMP.

It is well known that ageing is a risk factor for disease; common diseases of old age include cardiovascular diseases, neurodegenerative diseases and cancer [17]. In consideration of the background of the patient—whether they have other comorbidities, as is likely with increasing age—more stringent toxicity management may need to be applied. On the other hand, the patient’s medical condition may make them unable to tolerate harsh treatment regimens for the adverse events, such as steroids. Similarly, for paediatric studies, where additional ethical implications are imposed, and given the vulnerable nature of the population, investigators may consider a more conservative discontinuation criterion. Considering the age demographic of the study population and how this may impact the benefit/risk is important when standardising dose modification guidelines and any treatment recommendations by keeping the patient in the forefront of decisions. Additionally, each patient should be assessed holistically throughout implementation of toxicity management regimens with consideration for the impact on development, organ function, and reproduction as well as their ability to tolerate further interventions. The sponsor can reflect this through the TMG document to provide additional guidance to investigators.

Males and females can experience cancer incidence differently due to factors such as gene polymorphism, changes in enzymes involved in drug metabolism, hormones and other genetic factors. The sex of the patient is a factor that should be taken into consideration as studies have shown that women may have a lower capacity to clear certain drugs, such as fluorouracil, compared with men [18]. However, no differences in drug clearance were found between men and women for other drugs, like imatinib, indicating that the impact of sex on drug metabolism can vary widely depending on the specific medication [18]. It has also been shown that women can experience ADRs more frequently [19], with a study reporting that 11 drugs showed greater than 40% difference in PK measures between males and females, yet no dosing recommendations to consider sex were issued, based on the assumption that these differences were not clinically relevant [20]. Therefore, the sex demographic of the study population is an important element to reflect on when creating standardised toxicity management guidelines to ensure a conservative approach is maintained that keeps all patients protected.

In oncology studies, contraception is usually mandated to prevent pregnancy as embryo-foetal toxicity is not typically a characterised risk. It is safer for patients to avoid pregnancy when taking oncology drugs due to the potential for them to be highly toxic and lead to serious and severe adverse events. However, in non-oncology studies, contraception is not typically mandated but pregnancy and fertility are factors that should be considered when creating TMGs to ensure that best efforts are made for pregnancy prevention and protection against loss of fertility.

The standardised TMGs must be reviewed and updated regularly as medical knowledge progresses and advances. Institutional guidelines (from organisations such as American Society of Clinical Oncology [ASCO] and European Society for Medical Oncology [ESMO]), industry knowledge, and healthcare technologies are constantly evolving; therefore, there is a need for clear guidance where there is a difference in approach to, for example, the class or duration of treatment expected depending on posology/dosing regimen. A prevalent example of this are recent updates in guidelines surrounding prophylaxis like anti-emetics and granulocyte colony-stimulating factor (G-CSF) to manage nausea and neutropenia class effects, respectively [10, 21]. Consulting and incorporating perspectives from clinicians, regulatory agencies, and patient advocacy groups is worth considering in the development of these guidelines as it can further enhance their applicability and acceptance.

To approach this, adoption of standardised guidelines should accompany a periodic review cycle (e.g. annual), during which best practice guidelines are routinely consulted to confirm they are aligned with current best clinical practice. Ad hoc changes are advised if a finding is profound enough to justify a change to the standard template outside of the annual review period (e.g. mandating prophylactic use of G-CSF (granulocyte-colony stimulating factor) when the incidence of higher grade neutropenia is markedly greater than expected). Where considered appropriate, an expert knowledge group or individual for the relevant organ should be consulted as well as cross-functional consultation with the relevant stakeholders. It may also be worth considering incorporating perspectives from clinicians, regulatory agencies, and patient advocacy groups in the development of these guidelines to further enhance their applicability and acceptance. Alignment and endorsement across these functions, informed by experts, will ensure successful implementation and increased impact of the standardised TMGs across clinical trials.

Looking forward, as medical technology advances, there may be potential for the implementation of a periodic update process to utilise artificial intelligence (AI) to monitor the regulatory landscape for updates to best clinical practice guidelines [22]. For instance, natural language processing could be utilised to extract, interpret, and summarise complex medical documents to increase efficiency in key data extraction, summarisation, and communication of updates. and, even better, real-time alerts to patient safety teams could accompany this. Furthermore, AI-driven systems could streamline the integration of diverse data sources, such as clinical trial results and expert consensus, to provide a cohesive view of evolving best practices. By leveraging predictive analytics, AI can forecast potential changes in clinical guidelines, allowing teams to proactively adapt their guidelines. Additionally, personalised update recommendations tailored to specific specialties and practice environments could enhance the relevance and applicability of the information in the guidelines if AI was used as a tool alongside them. Overall, the strategic use of AI in this context promises to significantly enhance the timeliness and accuracy of implementing standardised TMG updates, ultimately improving patient outcomes.