Introduction: Additional considerations are required for the benefit-risk assessment of new drugs or indications in the setting of (neo)adjuvant cancer treatment as compared to the metastatic/advanced setting, possibly leading to different decision patterns for the (neo)adjuvant versus the metastatic and advanced setting within a health authority but also among different health authorities. Methods: We analyzed regulatory decisions at the Swiss Agency for Therapeutic Products Swissmedic (SMC) for all oncology indications (mostly metastatic indications) and indications in the (neo)adjuvant setting and compared these to decisions taken by the European Medicines Agency (EMA) and the United States Food and Drug Administration (FDA). Results: Comparing the positive and negative decisions within SMC between July 2017 and December 2021, the approval rates were with 66.7% lower for (neo)adjuvant indications versus 88.4% in the metastatic and advanced indications. While the approval rates for metastatic and advanced new active substances (NAS) applications were similar at SMC as compared to the EMA and the FDA, they were lower for (neo)adjuvant applications at SMC as compared to the EMA and the FDA. The underlying reason in all cases with divergent decisions at SMC as compared to EMA and FDA was that no overall survival benefit as compared to control arm has been observed in the submitted data package. Conclusion: Approval and consensus decision rates at SMC in comparison to EMA and FDA were lower for (neo)adjuvant indications but not for advanced and metastastic NAS oncology indications.

For a market authorization, therapeutic products including drugs are assessed by the respective regulatory agency. A positive benefit-risk ratio must be established by demonstrating that the benefits outweigh the risks in a given medical indication. The benefit-risk assessment has to be specific to the submitted indication, and the considerations can vary among different disease settings. From a regulator’s perspective, the benefit-risk assessment of a new drug in oncology depends on the clinical setting: curative or palliative. In the curative setting following resection, the patient is considered to be free of tumor, and a (neo)adjuvant treatment is administered before or after surgery to prevent recurrence and death, therefore increasing the proportion of cured patients. Despite the risk of recurrence, a substantial subset of patients is generally cured without any further (neo)adjuvant treatment. This is in contrast to the metastatic setting where all patients with few exceptions have an incurable tumor and a given treatment has a potential effect on the tumor of every single treated patient. These different clinical settings require different considerations for the benefit-risk assessment as outlined by us in a recently published comment [1]. In this article, we analyzed decision patterns at the Swiss Agency for Therapeutic Products Swissmedic (SMC) in the (neo)adjuvant versus the metastatic and advanced setting and compared these decisions to those of foreign regulatory agencies.

We compared regulatory decisions at SMC for all oncology indications (n = 69; including metastatic, hematological neoplasia, and one (neo)adjuvant indication) to indications in the (neo)adjuvant setting (n = 18). For the all oncology indications only new active substances (NAS) that received a final decision by SMC between 2009 and 2018 were included. For the (neo)adjuvant cancer setting, NAS and indication extensions (IEs) submitted to SMC between July 2017 and December 2021 were evaluated. All of these submissions were additionally compared to decisions taken by the European Medicines Agency (EMA) and the United States Food and Drug Administration (FDA).

The approval rates at SMC were lower for (neo)adjuvant indications as compared to all oncology applications with rates of 66.7% versus 88.4% (shown in Fig. 1). While the approval rates for all oncology applications among SMC, EMA, and FDA were similar [2], the approval rate for (neo)adjuvant applications at SMC was lower compared to those at the EMA (88.9%) and the FDA (94.4%). The consensus decision rate (common positive and negative decision among the respective agencies) for the (neo)adjuvant indications was 77.8% (14/18) between SMC and EMA and 72.2% (13/18) between SMC and FDA. Similar to the approval rates, consensus decision rates among all three agencies were similar for all oncology indications, while the consensus decision rate for (neo)adjuvant indications was lower between SMC and EMA and SMC and FDA in comparison to the consensus rate between EMA and FDA (94.4% [17/18]) (shown in Fig. 2). To understand the reasons for the lower approval rate for (neo)adjuvant indications, we evaluated the (marketing authorization applications) MAAs that received a divergent decision at SMC as compared to at least one agency that granted approval (shown in Fig. 3; Table 1).

Fig. 1.

Approval rates at SMC, EMA, and FDA for all oncology applications and for (neo)adjuvant indications.

Fig. 1.

Approval rates at SMC, EMA, and FDA for all oncology applications and for (neo)adjuvant indications.

Close modal
Fig. 2.

Consensus decision rates among the 3 agencies for all oncology applications and for (neo)adjuvant indications.

Fig. 2.

Consensus decision rates among the 3 agencies for all oncology applications and for (neo)adjuvant indications.

Close modal
Fig. 3.

Flow diagram with decisions for the (neo)adjuvant applications from the SMC perspective.

Fig. 3.

Flow diagram with decisions for the (neo)adjuvant applications from the SMC perspective.

Close modal
Table 1.

Divergent decisions

Tumor indicationDrug classPrimary end pointPrimary endpoint validated*Toxicity in the experimental arm (≥G3 AE, SAE, grade 5 AE +/− imAE)SMC approvalEMA approvalFDA approvalMain uncertainties at SMC
Melanoma; stage II Immune checkpoint inhibitor RFS No Increased No Yes Yes No OS analysis submitted 
Breast cancer; HER2 neg. CDK 4/6 inhibitor iDFS No Increased No Yes Yes OS follow-up short with too few events 
OS detriment cannot be ruled out 
Breast cancer Immune checkpoint inhibitor EFS/pCR No Increased No Yes Yes OS follow-up short with too few events 
Overlapping survival curves until occurrence of the majority of censorings 
Renal cell cancer Multikinase inhibitor DFS No Increased No No Yes No OS benefit at final analysis 
Renal cell cancer Immune checkpoint inhibitor DFS No Increased No Yes Yes OS follow-up short with too few events 
Overlapping survival curves until occurrence of the majority of censorings 
Tumor indicationDrug classPrimary end pointPrimary endpoint validated*Toxicity in the experimental arm (≥G3 AE, SAE, grade 5 AE +/− imAE)SMC approvalEMA approvalFDA approvalMain uncertainties at SMC
Melanoma; stage II Immune checkpoint inhibitor RFS No Increased No Yes Yes No OS analysis submitted 
Breast cancer; HER2 neg. CDK 4/6 inhibitor iDFS No Increased No Yes Yes OS follow-up short with too few events 
OS detriment cannot be ruled out 
Breast cancer Immune checkpoint inhibitor EFS/pCR No Increased No Yes Yes OS follow-up short with too few events 
Overlapping survival curves until occurrence of the majority of censorings 
Renal cell cancer Multikinase inhibitor DFS No Increased No No Yes No OS benefit at final analysis 
Renal cell cancer Immune checkpoint inhibitor DFS No Increased No Yes Yes OS follow-up short with too few events 
Overlapping survival curves until occurrence of the majority of censorings 

AE, adverse events; SAE, serious adverse events; imAE, immune-mediated adverse events; RFS, recurrence-free survival; OS, overall survival; iDFS, invasive disease-free survival; EFS, event-free survival; pCR, pathological complete response; DFS, disease-free survival.

*To correlate with OS (in tumor type AND the drug class).

In this article, we evaluated differences in decision patterns for the (neo)adjuvant versus the metastatic and advanced setting within SMC and compared these to decisions taken by the EMA and the FDA. Approval and consensus decision rates at SMC in comparison to EMA and FDA were lower for (neo)adjuvant indications but not for all, in particular metastastic, NAS oncology indications. Five MAAs in the (neo)adjuvant setting received a divergent decision at SMC as compared to at least one agency that granted approval. The underlying reason in all 5 cases with divergent, i.e., negative (=either rejection of the MAA or withdrawal by applicant), decisions at SMC was that no overall survival (OS) benefit as compared to control arm has been observed in the submitted data package. In 4/5 cases, no OS data were submitted or the OS follow-up was too short with few events at the time of assessment with overlapping survival curves until occurrence of the majority of censorings. In the (neo)adjuvant setting, a longer follow-up is usually needed to detect an OS difference. Therefore, a surrogate parameter is often chosen as primary end point in (neo)adjuvant trials (shown in Table 1). Acceptance of a surrogate end point is critically dependent on whether it has been validated with regard to OS in the given tumor type and stage for the drug class investigated [3, 4]. The surrogacy between primary end point and OS was not validated in any of the MAAs with a negative decision at SMC. Along with the uncertainty of an OS benefit or even the absence of an OS benefit at the final analysis and the increased toxicity in the experimental arm, the benefit-risk was negative for all of these MAAs. In this report, we followed up on these applications with previous negative SMC decisions until ASCO 2023. To the best of our knowledge, no OS updates were published or presented at congresses which would have been indicative of a clinically relevant OS benefit. Still, for most of the applications, the pivotal studies are ongoing for OS, and it cannot be ruled out that the benefit-risk might change over time with a longer follow-up period. For MAAs with high uncertainties for an OS benefit but with a statistically significant and clinically relevant therapeutic benefit in a surrogate endpoint, a temporary authorization might be useful. The temporary authorization pathway was introduced in Switzerland in 2019, but only for NAS. Most MAAs with (neo)adjuvant indications are IEs of already approved drugs. The possibility for a temporary authorization for IEs was introduced in 2023 [5]. This temporary authorization pathway will be important for the registration of (neo)adjuvant treatments. It will allow OS data to mature, which can then be used for a final benefit-risk assessment to either convert a temporary into a regular approval or revoke the temporary approval.

An ethics statement was not required for this study type, no human or animal subjects or materials were used.

The authors have no conflicts of interest to declare.

This study was not supported by any sponsor or funder.

Concept and design of the research, interpretation of data, and drafting the manuscript: M.Z-P. and U.-P.R.; data acquisition and analysis: M.Z-P., Q.L., S.J., A.G., and A.W.; and review and approval of the manuscript: M.Z-P., Q.L., S.J., A.G., A.W., and U.-P.R.

Due to legal grounds, the data used for this evaluation are not publicly available. Further inquiries can be directed to the corresponding author.

1.
Pavic
M
,
Wolfer
A
,
Li
Q
,
Gircys
A
,
Juritz
S
,
Wicki
A
, et al
.
Regulatory challenges: is a surrogate end point instead of overall survival enough for regulatory approval of (Neo)Adjuvant cancer treatment? The swissmedic perspective
.
J Clin Oncol
.
2023
;
41
(
32
):
4973
4975
. doi: .
2.
Rohr
UP
,
Iovino
M
,
Rudofsky
L
,
Li
Q
,
Juritz
S
,
Gircys
A
, et al
.
A decade comparison of regulatory decision patterns for oncology products to all other non-oncology products among Swissmedic, European Medicines Agency, and US Food and Drug Administration
.
Clin Transl Sci
.
2023
;
16
(
9
):
1569
81
. doi: .
3.
Merino
M
,
Kasamon
Y
,
Theoret
M
,
Pazdur
R
,
Kluetz
P
,
Gormley
N
.
Irreconcilable differences: the divorce between response rates, progression-free survival, and overall survival
.
J Clin Oncol
.
2023
;
41
(
15
):
2706
12
. doi: .
4.
Gyawali
B
,
Prasad
V
.
Making adjuvant therapy decisions with uncertain data
.
Ann Oncol
.
2019
;
30
(
3
):
361
4
. doi: .
5.
Swissmedic
.
Expansion of scope of temporary authorisations: temporary additional indications
will also be possible from 1 January 2023. 2022. URL:
https://www.swissmedic.ch/swissmedic/en/home/humanarzneimittel/authorisations/information/erweiterung-geltungsbereich-befristete-zulassungen.html