Measuring the value of new kidney cancer treatments using ASCO and ESMO scoring systems

Annually, an estimated 74,000 people are diagnosed with cancers of kidney and renal pelvis and almost 15,000 people die from these diseases in the Unites States [1]. Kidney cancer is the 10th most common cancer and the 8th most common cause of cancer death [2]. Renal cell carcinoma (RCC) accounts for 85% of kidney cancer and approximately 70% of patients with RCC have clear cell type. The remains consist of papillary, chromophobe and collecting duct tumors [3]. About a tenth of patients with RCC express advanced metastatic disease at the time of diagnosis. About a third of patients who undergo surgical resection for local disease experience recurrence, and need systemic treatment as well [4]. When surgical resection is feasible, it is the treatment of choice in loco-regional RCC. During 1980s and 1990s, cytokine therapy with interferon alpha or interleukin-2 was the systemic treatment available for unresectable or metastatic RCC. Since the 2000s, tyrosine kinase inhibitors (TKIs) and immune checkpoint inhibitors (ICIs) have been shown to be effective in patients with RCC, and various combination therapies have recently been introduced [5, 6]. Although the rapid development of novel therapies has provided insights into the future direction of treatments for RCC, the best choice of drug, and the sequence of drug remain elusive. Currently, due to the rapid development and approval of novel anticancer therapies, the high cost of cancer treatment has become a major concern for patients and the society. The financial toxicity may lead to psychosocial distress, poor quality of life (QoL), and worse patient outcomes. Unfortunately, increased costs do not always correlate with improved patient outcomes. Thus value-based decision-making is critical and includes clinical benefit of the drug, potential toxicities affecting QoL, and the cost-effectiveness [7, 8]. The American Society of Clinical Oncology (ASCO) and the European Society for Medical Oncology (ESMO) independently developed assessment tools to evaluate the value of new anticancer drugs and to facilitate decision-making of patients, doctors and payers [9, 10]. The ASCO value framework (ASCO VF) and ESMO-magnitude of clinical benefit scale (ESMO-MCBS), were released in 2015, ASCO VF was revised in 2016 (v2.0), and ESMO-MCBS in 2017 (v1.1) [11, 12].

This study aims to determine whether value assessment helps clinicians select the right anticancer drugs for RCC and whether they are suitable for social valuation. For the purpose, we review the value assessment tools (ASCO VF and ESMO-MCBS), and evaluate the value of emerging therapeutics in RCC.

Phase 3 clinical trials showing positive results of first-line therapy compared with sunitinib in patients with metastatic RCC and approved by United States Food and Drug Administration (US FDA) were evaluated. We searched relevant literature from PubMed and US FDA approval announcements. Two medical oncologist (HH, LH) independently assessed the eligibility of the study. After reviewing US FDA approval announcement and the trials supporting the approval decision, 6 trials were selected. We then retrieved additional data on patient-reported outcomes (PRO) and follow-up (FU) reports via PubMed. Authors applied the ASCO VF v2.0 and the ESMO-MCBS v1.1 to the trials and compared the results. Disagreements between the authors were resolved by consensus based on further discussion. We were trained for the tools and had applied the tools to more than 10 phase 3 trials for previous study. The inter-rater reliability was achieved [13].

TKI including sunitinib has been the standard-of-care for metastatic RCC since mid-2000. In the past 10 years, the influx of novel medical treatments for RCC treatment has led to remarkable progress. Advances in oncology were associated with a high price tag. Therefore, RCC was accompanied by considerable economic burden not only patients but also for the health care system. At the same time, clinicians are contemplating which drugs should be administered and in what order. The value encompassing efficacy, toxicity, and cost-effectiveness has been clearly highlighted. This study reviewed the clinical application of ASCO VF and ESMO-MCBS for the novel therapeutics in RCC.

Pazopanib was approved for RCC by US FDA in 2009 [15]. The COMPARZ trial confirmed the non-inferiority of pazopanib compared with sunitinib as the first-line treatment in 2013 (median PFS 8.4 vs. 9.5 months; median OS 28.4 vs. 29.3 months) [16]. Patients receiving pazopanib manifested more adverse events such as altered hair color, weight loss, alopecia and abnormalities in liver function test, while toxicities affecting QoL, including hand foot syndrome, fatigue and grade 3 or 4 hematologic abnormality were more frequent in the sunitinib group. Among the trials included in this study, the health benefit of pazopanib according to ASCO VF was the lowest (11.0) due to the omission of clinical benefit score, and ESMO-MCBS yielded grade 4 which was the highest in form 2c.

Among the 5 trials including ICI, the trial for nivolumab and ipilimumab combination was reported first in 2018, two axitinib combination trial in 2019, and the other TKI plus ICI in 2021 [17]. The FU report including extended survival data and PRO was reported only for nivolumab and ipilimumab combination in 2019 [18]. The trial for nivolumab plus ipilimumab reported OS as a primary endpoint, while the others reported PFS as a primary endpoint, suggesting the high value of nivolumab and ipilimumab combination based on both ASCO VF (net health benefit 41.9) and ESMO-MCBS (grade 5). According to form 2a of ESMO-MCBS, grade 4 can be given in case of HR ≥ 0.7 and gain ≥9 months. Despite not-reached median OS, the FU report showed meaningful OS benefit over 9 months (HR 0.66, p < 0.0001). Thus, preliminary grade 4 was given to the combination of nivolumab and ipilimumab, and upgrade was done by the improvement in toxicity resulting final grade 5.

The trial of avelumab and axitinib was analyzed in two groups, including overall patients (ITT population) and patients with PD-L1-positive tumor [19]. The primary endpoints were PFS and OS in the PD-L1-positive group, and the secondary endpoint was PFS in ITT population. The US FDA approved avelumab in combination with axitinib for RCC based on PFS in ITT population. Based on form 2b of ESMO-MCBS, ITT population scored grade 3 (HR 0.69). FU data including OS or PRO has yet to be reported, which may alter the value later. Currently, avelumab is approved for Merkel cell carcinoma and urothelial carcinoma in Korea, but not for RCC.

The primary endpoints in the trial involving pembrolizumab plus axitinib were PFS and OS [20]. Significant longer PFS (HR 0.69, p < 0.001) and OS (HR 0.53, p < 0.0001) were identified in the trial involving pembrolizumab and axitinib. Due to the short FU time (median 12.8 months), the median OS has yet to be reported. ASCO VF requires only HR, while ESMO-MCBS requires both HR and absolute duration of the survival gain. Thus, the highest value was achieved with ASCO-VF (net health benefit 45.3), while ESMO-MCBS gave the final grade 3. Similar to avelumab plus axitinib, FU data would change the value of pembrolizumab and axitinib.

CLEAR and CheckMate 9ER were reported in 2021 and both trials showed improved PFS (primary endpoint) and OS (secondary endpoint) [21, 22]. They achieved upgrade in ESMO-MCBS due to early stopping. In these trials, FU data would change the value in terms of tail of curve, and QoL.

While ASCO VF and ESMO-MCBS have been used, further discussion about the appropriate group of patients for value assessment is required. The JAVELIN renal 101 trial showed differences in benefit for PD-L1-positive patients and ITT population. Occasionally, the primary and secondary endpoints are set in different group of patients, and the approval (US FDA or Korea FDA) is not always based on the primary endpoint. Sometimes, it is not clear that which group, according to endpoint or approval, is appropriate for the assessment. It was not clear whether OS was the best representative parameter. The primary goal of cancer treatment is to prolong survival, and therefore ASCO VF and ESMO-MCBS prioritized OS over PFS and gave more weight to OS. Even in the trial with OS as secondary or co-primary endpoint, the OS was used. However, the statistical design of the trial setting the endpoints is crucial and subsequent treatment may have limited benefit in OS. Further, a more recent study would gain unfavorable value. Generally, data including extended survival and PRO are reported later, thus additional points for tail of curve, palliation or QoL are unavailable. In particular, although QoL points are included in both tools, however they cannot be evaluated properly in recently reported trials. Furthermore, the additional points constitute a substantial proportion of the value. Axitinib and pembrolizumab achieved the highest value by ASCO VF (48.7) while grade 1 by ESMO-MCBS due to median OS was not reached, despite meaningful HR for death. It was also unclear whether “laboratory abnormality only” toxicity was insignificant. Is grade 4 laboratory abnormality including liver enzyme elevation, neutropenia and anemia is less important than grade 1 hair loss or fatigue? Hematologic toxicity is important in hematologic malignancy, and grade 4 neutropenia or thrombocytopenia in solid cancer is clinically significant as well. It could also reduce the quality of life due to recurrent transfusion and the risk of bleeding or infection. Thus uniform format which excludes “laboratory only toxicity” requires further discussion. Frequent errors occurred during toxicity analysis. However, the toxicity score was a relative measure of two arms, and the effect of error was marginal. Another shortcoming relates to the drug cost, which was not reflected adequately in the tools. The high cost is one of the major reasons for value assessment. ASCO VF has an entry for the cost, but it is only a reference. Further, the non-inferiority trial was not assessed appropriately. Pazopanib showed similar efficacy and a favorable toxicity profile compared with sunitinib and yielded the second highest grade [4] based on ESMO-MCBS but the lowest value (11.0) by ASCO VF. ESMO recommends the use of ESMO-MCBS for comparison of trials, while ASCO recommends against the use of VF for comparison. The discrepancy of results by the two evaluation tools may be due to the different purpose of the tools. Considering the purpose of “value evaluation of chemotherapeutics”, it is desirable to verify the results for the consistent results. These findings suggest that “similar efficacy and better toxicity” are more valuable than “better efficacy and similar toxicity”. Even allowing for the importance of PRO, the majority of oncologists appreciate the efficacy more than the toxicity. The results show that the highest score in one assessment tool does not mean the best treatment.

ASCO VF and ESMO-MCBS were applied to the new emerging therapeutics in RCC under the first-line setting and assigned the highest value cabozantinib plus nivolumab (net health benefit 49.2) and nivolumab plus ipilimumab combination (grade 5), respectively. These tools warrant further discussion and improvement. How should these assessment tools be applied in the real-world setting for approval, decisions of reimbursement or post-marketing assessment, and for patients, clinicians, decision-makers or payers (government or insurance company)? Appropriate application and revision of the tools requires in-depth discussion of experts in various fields.

Not applicable.

Conceptualization: Lee HY, Kang JH, Data curation: Ha H, Lee HY, Formal analysis: Ha H, Lee HY, Methodology: Kang JH, Software: Bae SJ, Validation: Kim DY, Investigation: Ha H, Lee HY, Writing-original draft: Ha H, Lee HY, writing – review & editing: Ha H, Lee HY, Kang JH, Bae SJ, Kim DY. The author(s) read and approved the final manuscript.

This work was supported by the INHA UNIVERSITY Research Grant.

The datasets generated and/or analyzed during the current study are available in the figureshare repository, 10.6084/m9.figshare.18585461.v1 If additional analysis data is needed, The datasets used and/or analyzed during the current study available from the corresponding author.