Anaïs Frappé
Partner
Windrose Consulting Group in London

Overall Survival (OS) has historically served as the “gold standard” endpoint in Health Technology Assessments (HTA) for oncology drugs: a definitive and objective measure capturing the ultimate goal of prolonging life.

Particularly, Overall Survival has been the crux of HTAs of medicines targeting incurable and/or metastatic cancers: often times, not demonstrating a clinically meaningful and statistically significant OS improvement vs. an accepted comparator limited HTA outcomes or led to decisions of no reimbursement, particularly in indications where treatment options are available to patients. Recently, it even led manufacturers to end their clinical development programs, anticipating difficult HTA discussions with payers.

Mattia Belloni
Senior engagement
Windrose Consulting Group

However, the cancer patients’ journey is complex and the benefit of a drug goes well beyond survival, even in cases of incurable and/or metastatic cancer. Non-OS endpoints can provide an extra lens through which to assess the benefits of an intervention, capturing dimensions such as time to disease-relevant events, patients’ response to the treatment, and other patient-centered metrics such as Quality of Life (QoL) and functional outcomes that OS alone does not fully reflect.

In this article, we focused our analyses on the EU-4 markets and recognize the central role of OS in oncology HTAs. At the same time, we highlight the importance of non-OS endpoints and advocate for a more consistent and systematic integration of these endpoints in HTA decisions.

The Central Role of Overall Survival in HTAs

While OS provides a definitive and objective measure and its role in evaluating a drug’s benefit – particularly in incurable and/or metastatic cancers – is undeniable, over indexing on OS carries limitations that can hinder the ability of payers to fully appreciate the therapeutic improvements of new treatments.

One of the primary drawbacks of OS is the need for extended follow-up periods to capture survival data. This necessitates longer trial durations, which can be impractical for slowly progressing or early-stage cancers. This has led some oncology treatments, now considered standard of care, to receive minimal HTA outcomes. A recent example is Krazati (adagrasib) in second line of KRAS G12C-mutated Non-Small Cell Lung Cancer (NSCLC), in which Bristol Myers Squibb requested an important SMR / ASMR IV to the French Transparency Committee (TC), on the basis of the KRYSTAL-12 ph3 study that compared adagrasib vs. docetaxel. In this case, the TC noted the lack of statistical significance in OS in an interim analysis, as well as the exploratory nature of the QoL data, and awarded a weak SMR and ASMR V. The statistically significant Progression-Free Survival (PFS) improvement (mPFS of 5.49 months in the adagrasib arm vs. 3.84 months in the docetaxel arm, HR 0.58, 95% CI, 0,45 ; 0,76, p < 0,0001) and Objective Response Rate (ORR) (31.9% vs. 9.2%, 95% CI, 2,56 ; 8,56, p < 0,0001) were not considered sufficient to counter-balance the lack of OS and QoL.

Additionally, OS is subject to confounding factors that complicate the ability to isolate the effect of the intervention. This can be the case when patients receive subsequent lines of treatments, or in randomized trials when crossover is allowed.

Last, while prolonging survival is the key goal of cancer treatments, OS does not capture other aspects of patient benefit, such as improvements in treatment response, QoL and functional outcomes, or delays in disease-relevant events, which are often considered equally important by physicians, patients, and caregivers.

Here, we recognize that payers do consider non-OS endpoints, as illustrated by the recent Krazati case in France, where the absence of QoL data limited HTA outcomes. However, we advocate for a more systematic and balanced recognition of these endpoints alongside OS, to ensure that broader dimensions of clinical value consistently inform HTAs.

Beyond Overall Survival: A Closer Look at Non-OS Endpoints

Beyond OS, a range of endpoints provide critical insights on a drug’s efficacy and impact on patients’ life. These endpoints can be segmented into three categories [see figure 1]:

• Time-to-Event Endpoints: endpoints that measure the time from treatment initiation to a specific event or outcome. Examples notably include Disease- / Event- / Metastasis- / Relapse-Free Survival (DFS, EFS, MFS, RFS), Progression-Free Survival (PFS), Progression-Free Survival 2 (PFS2), Symptom-Free Survival, Time to Deterioration (TTD), Time to Treatment Failure (TTF), or Time to Progression (TTP);
• Response-Based Endpoints: endpoints that assess the tumour response to a treatment, often based on imaging or clinical measurements. These may include Objective Response Rate (ORR), Complete Response (CR), Partial Response (PR), Disease Control Rate (DCS), Duration of Response (DoR), Minimal Residual Disease (MRD);
• Patient-Centered Endpoints: patient-reported endpoints that focus on patients' experience and well-being during and after treatment, including (Health-Related) Quality of Life ((HR)QoL).

Figure 1 – Non-OS Oncology Endpoints: Proposed Classification and Overview¹

These endpoints complement OS by capturing earlier signals of efficacy and offering a more nuanced outlook of how a therapy may improve patients’ daily living and alter the course of the disease. While our proposed categorization of endpoints intends to provide a framework for organizing non-OS endpoints into distinct segments, time-to-event and response to a treatment are often interrelated and some endpoints might span multiple categories. For example, Time to Deterioration (TTD), which measures the time to a clinically meaningful worsening in symptoms or performance status is both time- and patient-related.

The Opportunity of Non-OS Endpoints in HTA Outcomes

Figure 2 – Kadcyla’s Summary of EMA Registrational Data in HER2+ Early Breast Cancer⁵

Figure 3 – Kadcyla’s Summary of HTA Outcomes in the EU-4 in HER2+ Early Breast Cancer⁶

Payers increasingly recognize the importance of non-OS endpoints in their assessments of cancer treatments, particularly in the curative setting, thus balancing timely patient access to innovative therapies with the need for robust evidence.

Figure 4 – Opdivo Summary of EMA Registrational Data in Oesophageal or Gastroesophageal Junction Cancer⁷

For example, Kadcyla (trastuzumab emtansine) obtained its European Medicine Agency (EMA) Market Authorization (MA) for the “Adjuvant treatment of adult patients with HER2-positive early breast cancer who have residual invasive disease, in the breast and/or lymph nodes, after neoadjuvant taxane-based and HER2-targeted therapy” in November 2019, on the basis of the KATHERINE ph3 open-label clinical trial, notably based on a significant improvement in its primary endpoint of Invasive-Disease-Free survival (IDFS) vs. Herceptin (HR 0.50; 95% CI, 0.39-0.64; p <0.001), where IDFS occurred in 91 patients who received Kadcyla (12.2%) and 165 patients who received Herceptin (22.2%), and on additional secondary endpoints that included IDFS(in this case also considering second primary non-breast invasive cancers),² DFS,³ and OS.⁴ OS was immature at the time of EMA approval, and subsequent payer submissions. However, Kadcyla achieved reimbursement in the EU-4 markets, notably achieving SMR Important / ASMR III in France, and minor Added Benefit in Germany. All four payers recognized the IDFS / DFS improvements, shadowing the immature OS data, and somewhat unfavourable safety profile [see figures 2, 3].

Figure 5 – Opdivo Summary of HTA Outcomes in the EU-4 in Oesophageal or Gastroesophageal Junction Cancer⁸

However, while these endpoints are generally accepted in the curative oncology space, payers continue to require a demonstrated OS improvement. In addition, the relevance of these endpoints and the evidence requirements have not always been consistent across the the EU-4, leading to discrepancies in HTA outcomes. Indeed, in Oesophageal / Gastroesophageal Junction Cancer, Opdivo (nivolumab)achieved favourable HTA outcomes in France, Germany, and Spain for the “adjuvant treatment of adult patients with oesophageal or gastro-oesophageal junction cancer who have residual pathologic disease following prior neoadjuvant chemoradiotherapy as monotherapy” based on superior DFS and in spite of immature OS data, but was considered non innovative and denied reimbursement in Italy, where AIFA considered that “Since the greater efficacy was demonstrated in terms of DFS and PFS2, but not in terms of OS, the only clinically relevant endpoint in this setting, the added therapeutic value of nivolumab in this indication can be considered poor” [see figures 4 and 5].

Conclusion

While OS shall remain the foundational endpoint in the assessment of medicines in oncology, we recommend a more consistent integration of non-OS endpoints to further support comprehensive value-based decision-making in oncology HTAs.

Sources

1. Adapted from the classification proposed by Clarivate in “Beyond overall survival: Time to agree on the value of alternative oncology endpoints?” (May 2024)
2. 95 (12.8%) in Kadcyla arm and 167 (22.5%) patients in Herceptin arm (HR 0.51; 95% CI, 0.40-0.66)
3. 98 (13.2%) patients with a DFS event in Kadcyla arm and 167 (22.5%) in Herceptin arm (HR 0.53; 95% CI, 0.41, 0.68)
4. 42 deaths (5.7%) in Kadcyla arm and 56 (7.5%) deaths in the Herceptin arm (HR : 0.70; 95% CI, 0.47-1.05)
5. EMA Assessment: https://www.ema.europa.eu/en/documents/variation-report/kadcyla-epar-h-c-2389-ii-0045-epar-assessment-report-variation_en.pdf; Clinical Trial: https://clinicaltrials.gov/study/NCT01772472; Trial Design: https://www.kadcyla-hcp.com/early-breast-cancer/efficacy/trial-information.html; EMA Label: https://www.ema.europa.eu/en/documents/product-information/kadcyla-epar-product-information_en.pdf; Outcomes: https://www.kadcyla.com/content/dam/gene/kadcyla/patient/pdfs/downloads/ebc_patient_brochure.pdf; https://www.nejm.org/doi/pdf/10.1056/NEJMoa1814017; https://www.kadcyla-hcp.com/early-breast-cancer/efficacy/trial-information.html; https://www.cadth.ca/sites/default/files/pcodr/Reviews2020/10182TrastuzumabEmtansineEBC_fnCGR_REDACT_EarlyConv_22Jan2020_final.pdf
6. IQWIG: https://www.g-ba.de/bewertungsverfahren/nutzenbewertung/515/#nutzenbewertung; G-BA: https://www.g-ba.de/bewertungsverfahren/nutzenbewertung/515/#English; HAS: https://has-sante.fr/jcms/p_3188463/en/kadcyla; AIFA: https://www.aifa.gov.it/documents/20142/1635388/12_Kadcyla.pdf; https://www.gazzettaufficiale.it/atto/serie_generale/caricaDettaglioAtto/originario?atto.dataPubblicazioneGazzetta=2021-09-28&atto.codiceRedazionale=21A05695; AEMPS: https://www.aemps.gob.es/informa/informes-de-posicionamiento-terapeutico/informe-de-posicionamiento-terapeutico-de-trastuzumab-emtansina-kadcyla-para-el-tratamiento-adyuvante-del-cancer-de-mama-precoz-her2-positivo/
7. EMA EPAR: https://www.ema.europa.eu/en/documents/variation-report/opdivo-h-c-3985-ii-0095-epar-assessment-report-variation_en.pdf; EMA label: https://www.ema.europa.eu/en/documents/product-information/opdivo-epar-product-information_en.pdf; Clinical trial: https://clinicaltrials.gov/study/NCT02743494
8. GBA: https://www.g-ba.de/downloads/91-1385-728/2022-02-17_Geltende%20Fassung_Nivolumab_D-728.pdf; https://www.g-ba.de/bewertungsverfahren/nutzenbewertung/728/#beschluesseI; QWIG: https://www.iqwig.de/download/a21-108_nivolumab_extract-of-dossier-assessment_v1-0.pdf; HAS: https://has-sante.fr/jcms/p_3324307/en/opdivo-nivolumab-cancer-oesophage-ou-jonction-oeso-gastrique; AIFA: https://www.gazzettaufficiale.it/atto/serie_generale/caricaDettaglioAtto/originario?atto.dataPubblicazioneGazzetta=2023-08-10&atto.codiceRedazionale=23A04506&elenco30giorni=true; https://www.aifa.gov.it/documents/20142/2001921/7_OPDIVO_esofago_adiuvante_Scheda_innovativita-GRADE.pdf; AEMPS: https://www.cofguadalajara.es/cofgu%5CDocumentos%5CRepositorios%5CNotas%20informativas%20AEM%5CIPT-154-Opdivo-nivolumab-cancer-esofago.pdf; https://botplusweb.farmaceuticos.com/FichaMUH/299332