Communication of anticancer drug benefits and related uncertainties to patients and clinicians: document analysis of regulated information on prescription drugs in Europe

Abstract Objective To evaluate the frequency with which relevant and accurate information about the benefits and related uncertainties of anticancer drugs are communicated to patients and clinicians in regulated information sources in Europe. Design Document content analysis. Setting European Medicines Agency. Participants Anticancer drugs granted a first marketing authorisation by the European Medicines Agency, 2017-19. Main outcome measures Whether written information on a product addressed patients’ commonly asked questions about: who and what the drug is used for; how the drug was studied; types of drug benefit expected; and the extent of weak, uncertain, or missing evidence for drug benefits. Information on drug benefits in written sources for clinicians (summaries of product characteristics), patients (patient information leaflets), and the public (public summaries) was compared with information reported in regulatory assessment documents (European public assessment reports). Results 29 anticancer drugs that received a first marketing authorisation for 32 separate cancer indications in 2017-19 were included. General information about the drug (including information on approved indications and how the drug works) was frequently reported across regulated information sources aimed at both clinicians and patients. Nearly all summaries of product characteristics communicated full information to clinicians about the number and design of the main studies, the control arm (if any), study sample size, and primary measures of drug benefit. None of the patient information leaflets communicated information to patients about how drugs were studied. 31 (97%) summaries of product characteristics and 25 (78%) public summaries contained information about drug benefits that was accurate and consistent with information in regulatory assessment documents. The presence or absence of evidence that a drug extended survival was reported in 23 (72%) summaries of product characteristics and four (13%) public summaries. None of the patient information leaflets communicated information about the drug benefits that patients might expect based on study findings. Scientific concerns about the reliability of evidence on drug benefits, which were raised by European regulatory assessors for almost all drugs in the study sample, were rarely communicated to clinicians, patients, or the public. Conclusions The findings of this study highlight the need to improve the communication of the benefits and related uncertainties of anticancer drugs in regulated information sources in Europe to support evidence informed decision making by patients and their clinicians.


Detailed coding scheme and strategy
Availability of information on drug benefits In our coding scheme for evaluating whether written product information addressed patients' general questions about the drug and the benefits they might expect, we examined the availability of information on the following items:

Broad domain
Specific categories of information What the drug is Type of medicine and name of the active substance What and who the drug is used for Indication and target population, including: • type of cancer; • stage of disease; • any special characteristics of the disease or patient (e.g., specific tumour markers, high/low risk disease, subgroups not eligible for treatment); • line of therapy; • sequencing of therapy; • specification of approved combination treatment How the drug works Drug's mechanism of action in the body The purpose of drug treatment Whether the drug is intended to cure disease, reduce the risk of disease, prevent the worsening of disease, or palliate disease symptoms How was the drug studied Number of main studies supporting approval and key aspects of study design, including • whether studies were: randomised controlled trials, non-comparative trials (including dosecomparator) randomised trials, single arm studies, or some other type of study design; • what treatment the experimental drug was compared to, if any; • how many patients participated the studies; • how the studies evaluated drug efficacy (the study endpoints) Benefits shown Study findings (included the nature and magnitude of any demonstrated benefits) with respect to: • pre-specified primary study endpoint(s); • overall survival; • health-related quality of life Our coding scheme did not include all categories of information listed in our initial taxonomy (Table 1).
Our initial taxonomy also included the category: 'onset and duration of drug benefit'that is, how long a patient should expect to wait before experiencing drug benefits, and how long those benefits would be expected to last. However, this information was not routinely available for all drugs in the sample so we did not include this category in the final coding scheme.
Availability of information on benefit uncertainties In our coding scheme for evaluating whether written product information addressed patients' concerns about the strength, clinical relevance and applicability of the available evidence with respect to drug benefits, we examined the availability of information on the following items: Broad domain Specific categories of information Strength/ quality of evidence (threats to internal validity) Regulatory concerns that studies may have yielded exaggerated treatment effects due to deficiencies in design, conduct and/or analysis, specifically: • whether studies were terminated early; • whether measurement of subjective study outcomes was unblinded; • whether studies suffered from missing outcome data; • whether studies lacked a control arm, or lacked an appropriate control; Clinical relevance of evidence Regulatory concerns that study findings may not translate to a meaningful clinical benefit because: • the magnitude of the effect was small/not clinically relevant; • studies included inappropriate endpoints or unvalidated, surrogate measures of drug efficacy Applicability of evidence (threats to external validity Regulatory concerns that study findings may not be generalisable to specific populations in clinical practice We also collected information on the regulatory and therapeutic characteristics of the sample, including whether drugs were granted a conditional marketing authorization or any special designation -for example, orphan designation, 'advanced therapy medicinal product' classification, or designation as a priority medicine under the EMA's PRIME scheme.
For drugs granted a conditional marketing authorisation, all SmPCs and Public Summaries should contain a statement acknowledging the conditional status of the market authorisation and making some reference to the additional postmarketing studies required for full approval. However, where this statement did not specify the nature of the evidence gap or uncertainties that postmarketing studies were meant to address, we did not consider the statement to have reported relevant uncertainties.

Detailed information on data sources
European Public Assessment Reports (EPARs) related to the initial marketing authorisation for all drugs in our sample were retrieved from the 'Initial marketingauthorisation documents' section of EMA's database of human medicines (https://www.ema.europa.eu/medicines).
Because the Summary of Product Characteristics (SmPCs), Public Summaries and Patient Information Leaflets (PILs) in EMA's database are updated following approval of line extensions and as new information becomes available, original SmPCs and PILs (matching the versions available at market entry) were obtained from the European Commission's Union Register of Medicinal Products for Human Use (https://ec.europa.eu/health/documents/community-register/html/), which contains a chronological listing of current and archived versions of the product information approved since product launch. There is no similar archive for Public Summaries. However, we were able to identify and extract the information in the Public Summaries relevant to each drugs' initial marketing authorisation.
Appendix Overall survival or quality of life included as secondary outcome Overall survival studied as a secondary outcome 27 (73.0) Quality of life studied as a secondary outcome 29 (78.4) a Denominator corresponds to the total number of studies (n=37) supporting the approval of 32 cancer drug indications during our study period. b These include one dose-comparator study, and one study where a combination of two experimental drugs was compared to one of the experimental drugs alone. "Patients treated with Zejula lived on average 11.3 months without their disease getting worse compared with 4.7 months in patients treated with placebo (a dummy treatment)... This may allow the next cycle of platinum-based therapy to be delayed."

Comment:
• "Patients taking Fotivda lived for longer without their disease worsening (12 months) than those given another approved medicine sorafenib (9 months)."

Comment:
• "The main measure of effectiveness was response to treatment. Patients were considered to have responded if they had no remaining cancerous B cells in their blood and bone marrow after treatment. An analysis of the first 218 patients treated showed that after at least 2 cycles of treatment, 81% (88 out of 109) of patients receiving Besponsa responded to treatment compared with 29% (32 out of 109) of patients receiving other chemotherapy. Patients who responded to treatment could proceed to have a stem cell transplant."

Comment:
• Explains 'response rate' in lay terms • Explains what this means in terms of a clinically relevant outcome Alectinib (Alecensa) "Response to treatment was assessed using body scans and standardised criteria for solid tumours, with complete response being when the patient had no remaining signs of the cancer. In the first study around 52% of patients given Alecensa (35 out of 67) were considered by the treating doctors to have shown a complete or partial response to the medicine at the time of analysis. In the second study, the complete or partial response rate at the time of analysis was 51% (62 out of 122 patients). Response was maintained for an average of 14.9 months in the first study, and 15.2 months in the second study."

Comment:
• Incomplete explanation of response rate. Describes 'complete response' in lay terms, but no explanation of 'partial response' • No explanation of what this might mean in terms of a clinically relevant outcome Midostaurin (Rydapt) "Overall, using the most stringent, up-to-date criteria, the disease responded to treatment in about 28% of patients (32 of 113). When looked at separately the response rate was highest (60%) in those with aggressive systemic mastocytosis." Comment:

• No explanation of what 'responding to treatment' means • No explanation of what this means in terms of a clinically relevant outcome
Appendix Box 1. Communication of mechanism of action in PILs Alectinib (Alecensa) for ALK-positive advanced NSCLC: • "Alecensa blocks the action of an enzyme called 'ALK tyrosine kinase'. Abnormal forms of this enzyme (due to fault in the gene that makes it) help encourage cancer cell growth. Alecensa may slow down or stop the growth of your cancer. It may also help to shrink your cancer." Avelumab (Bavencio) for metastatic Merkel cell carcinoma: • "PD-L1 is found on the surface of MCC cells, and helps protect tumour cells from the immune system (the body's natural defences). Bavencio binds to PD-L1, and blocks this protective effect, allowing the immune system to attack the tumour cells." Lutetium (177Lu) oxodotreotide (Lutathera) for unresectable or metastatic somatostatin receptor positive GEP-NETs: • "The tumour needs to have somatostatin receptors on the surface of its cells in order for the medicine to be effective. Lutathera binds with these receptors and emits radioactivity directly into the tumour cells, causing their death." Tisagenlecleucel (Kymriah) for relapsed or refractory DLBCL: • "The T cells are taken from your blood and a new gene is put into the T cells so that they can then find the cells causing your cancer. When Kymriah is infused into your blood, the modified T cells will find the cancer cells and destroy them." Lorlatinib (Lorviqua) for ALK-positive advanced NSCLC: • "Lorviqua inhibits a type of enzyme called tyrosine kinase and triggers the death of cancer cells in patients with alterations in genes for ALK. Lorviqua is only given to patients whose disease is due to an alteration in the gene for ALK tyrosine kinase." Cemiplimab (Libtayo) for metastatic or locally advanced cutaneous squamous cell carcinoma: • "LIBTAYO works by helping your immune system fight your cancer." 12 Appendix lack of convincing comparative data, lack of generalizability) preclude any conclusion about the efficacy of rucaparib compared to trabectedin+PLD. According this view, there are significant concerns about potential loss of efficacy in the absence of more definitive and direct comparative data. Furthermore, in many instances these patients would likely have been treated with PARP inhibitors in the maintenance setting making the target population somewhat theoretical. There were also concerns that patients could be misguided by apparent advantages in toxicity without due consideration to uncertainty about efficacy" (p.112).
Divergent opinion with EMA's Scientific Committee: "Based on the currently available data we consider the benefit-risk balance for rucaparib in the proposed restricted indication to remain undetermined. Major uncertainties exist regarding the quality of the data, Internal inconsistency in outcomes "There is internal inconsistency in the outcomes, as the isolation of the measured effect to hormone receptor positive patients lacks a clear rationale, contributing to uncertainty" (p.147). patients, is severe in a high proportion, and can be expected to affect quality of life… in the context of deficiencies in the efficacy demonstration, the adverse effect profile of neratinib is a matter of important concern" (p169). "Lorviqua was also effective when the cancer had spread to the brain. Depending on which previous treatment the patients had received, "Lorviqua inhibits a type of enzyme called tyrosine kinase and triggers the death of cancer cells in patients with alterations in genes for ALK. Lorviqua is only given to patients whose disease is due to an alteration in the gene for ALK tyrosine kinase." (Emphasis added)

Information from sections 'What benefits of drug X have been shown in studies?' and "Why is drug X approved?' in Public Summaries
Information from section 'How does drug X work?' in PILs a comparator; however, the ability to induce complete remission, although in a very small fraction of subjects, is noted" (p105).
Need for additional data in second line use "Only 28 patients were included in the cohort EXP-3B including patients who have progressed after a second generation ALK inhibitors used as a first line treatment. As a consequence further data are needed to confirm the efficacy of lorlatinib in that setting and the applicant agreed to conduct a prospective observational single arm study to confirm the observed results from EXP-3B" (p107).
Divergent opinion within EMA's Scientific Committee "Major therapeutic advantage has not been shown for the entire claimed indication, as efficacy in the secondline setting (i.e. after a previous treatment with a second-generation ALK tyrosine kinase inhibitor) is around 67% and 52% of patients treated with Lorviqua had no signs of cancer in the brain or the signs of cancer had reduced." "Lorviqua has been given 'conditional authorisation'. There is more evidence to come about the medicine, which the company is required to provide."

Information from section 'How does drug X work?' in PILs
currently not sufficiently established. profile, albeit the number of exposed patients is low. In conclusion, evidence-based clinical decisions to use larotrectinib are only justified for the rare conditions listed above and in situations where established alternatives are lacking or, as in the case of major surgical procedures, where available alternatives are associated with high morbidity and mortality. From a clinical decision perspective, use of larotrectinib may be an attractive therapeutic option when established effective treatments are lacking and when based on relevant tumour tissue sequencing one could confirm the presence of NTRKfusions and exclude other known oncogenic "drivers". However, such decisions cannot be considered evidence-based due to the lack of clinical evidence and the lack of predictive ability of clinical decision algorithms purely based on sequencing data. Nevertheless, such approaches warrant further investigation" (p.150).

Information from sections 'What benefits of drug X have been
shown in studies?' and "Why is drug X approved?' in Public Summaries

Information from section 'How does drug X work?' in PILs
Uncertainty around estimates of benefit due to methodological deficiencies "The application is considered lacking in prospectively studied cohorts that could provide an unbiased estimate of [overall response rate]. The uncertainty about the magnitude of the effect estimate due to these circumstances is of importance with respect to SmPC claims, and further strengthens the case for the noncomprehensiveness of available data. A prospective cohort is required to produce an unbiased estimate of efficacy" (p188).
"The non-randomised design further hampers the assessment of particularly the time-dependent outcomes. The small efficacy data base raises issues with regard to the representativeness in relation to the indication sought, encompassing any solid tumour type. This aspect will be to some extent addressed during the post-marketing study and external controls. This

Information from sections 'What benefits of drug X have been
shown in studies?' and "Why is drug X approved?' in Public Summaries

Information from section 'How does drug X work?' in PILs
uncertainty is stated in the SmPC" (p188).
"Due to the small sample size, the confidence intervals are generally wide, making efficacy estimates generally imprecise and hampering the possibility to draw conclusions regarding efficacy in subgroups, e.g. with regard to age groups and gene fusion type. This aspect will be addressed by the post-marketing study)" (p188).
"Overall, notwithstanding the considerable methodological caveats outlined above, the efficacy estimates available today may be considered outstanding in this generally late stage disease setting. The main issue efficacy-wise is the robustness and generalisability of these estimates. While it is likely that the estimates may change, possibly in a negative direction, the present outstanding estimates provide some reassurance as to the presence of a large treatment Need for additional data in first line "…although the data in 1st line was promising, the data is limited as few patients have reached the 6 month milestone and further confirmatory data from additional treated patients would be needed to confirm the effect size of the benefit" (p89, also see p90). There is concern that associated fibrosis may make surgery more difficult and impact post-operative wound healing although the limited available data does not indicate greater difficulty to perform radical prostatectomy" (p139).
Expert consultation: "Tookad VTP is claimed to primarily defer rather than avoid the need for radical therapy... and in the short term there seemed to be a positive benefitrisk balance. However, in the context of low-risk prostate cancer, with active surveillance being a valid management option and the existence of effective average progression occurred after 28 months in patients treated with Tookad compared with 14 months in those who did not receive treatment." "It is unclear… what effect Tookad treatment will have on future treatments for prostate cancer and more study is needed to find out if such treatments are compromised." limited and so, at this time, data are currently not available to know whether the benefit of TOOKAD-VTP is long-lasting. If you do require further treatment, at the moment, there is limited information on whether TOOKAD-VTP affects the efficacy and safety results of other treatments (such as surgery to remove the prostate or radiotherapy)."

Drug Indication
Nature of EMA concerns about study methods and findings as reported in EPAR Information from sections 'What benefits of drug X have been shown in studies?' and "Why is drug X approved?' in Public Summaries Information from section 'How does drug X work?' in PILs treatments in case of progression, it is the long-term benefits (and risks), which bear more weight in the overall benefit-risk assessment… [T]he follow-up of available studies with Tookad VTP is of insufficient duration to understand the long-term effects given the complex and heterogeneous prostate cancer biology requiring 10 or more years of followup. Without this information, it is impossible to make informed treatment decisions… For a majority of Scientific Advisory Group members, this risk was considered unjustified, in the absence of efficacy data showing at least non-inferiority of the approach in terms of long-term outcome. Given the existence of highly effective alternative therapy (radical therapy) and the fact that the disease is not rare, a risk of compromising long-term outcome was not considered justified" (p97).
Divergent opinion within EMA"s Scientific Committee:

Drug Indication
Nature of EMA concerns about study methods and findings as reported in EPAR Information from sections 'What benefits of drug X have been shown in studies?' and "Why is drug X approved?' in Public Summaries

Information from section 'How does drug X work?' in PILs
"Although a short term benefit in terms of progression of the disease has been shown in study PCM301, the clinical relevance remains to be established… [T]he efficacy data are not sufficient to support at least noninferiority in terms of long-term outcomes… Importantly, there is virtually no data on the potential consequences of post-Tookad local scarring in case of disease progression. Thus the safety and efficacy of subsequent radical therapy is uncertain and it still needs to be demonstrated that treatment with Tookad does not compromise the eligibility for and the results of subsequent radical therapy" (p145). analysis, lack of convincing comparative data, lack of generalisability) preclude any conclusion about the efficacy of rucaparib compared to trabectedin+PLD… [T]here are significant concerns about potential loss of efficacy in the absence of more definitive and direct comparative data. Furthermore, in many instances these patients would likely have been treated with PARP inhibitors in the maintenance setting making the target population somewhat theoretical" (p112).

Rucaparib
Divergent opinion within EMA's scientific committee: "Based on the currently available data we consider the benefit-risk balance for rucaparib in the proposed restricted indication to remain undetermined. Major uncertainties exist regarding the quality of the data, as the current pivotal results for the rucaparib application are derived from a post-hoc analysis from pooled data platinum based medicines, 65% (51 patients) had a response to treatment with Rubraca and the response lasted on average 294 days (around 10 months)." "The European Medicines Agency decided that although further study was needed to better understand the size of the benefit, Rubraca's benefits are greater than its risks and it can be authorised for use in the EU." "Rubraca has been given 'conditional authorisation'. This means that there is more evidence to come about the medicine, which the company is required to provide." DNA in the cancer cells, resulting in their death." (Emphasis added)