ASCO, ESMO Value Frameworks Imperfectly Apply to Hematology

Current value frameworks issued by leading medical societies may not be easily generalizable to the treatment of hematologic malignancies, according to two studies presented at the 2016 ASH Annual Meeting and Exposition.

The American Society of Clinical Oncology (ASCO) and European Society of Medical Oncology (ESMO) have issued value frameworks in recent years to combat the rising costs associated with cancer treatment. However, these guidelines were primarily derived from data on the treatment of solid tumors, and their applicability to hematologic cancers remains unclear.

Matthew C Cheung, MD, SM, FRCPC, associate scientist at Sunnybrook Health Sciences Centre’s Odette Cancer Centre (Toronto, ON), and colleagues conducted a retrospective review of randomized controlled trials (RCTs) investigating new therapies in hematologic malignancies to determine if they met the threshold for value by ASCO and/or ESMO standards. Erlene K Seymour, MD, hematologist at Barbara Ann Karmanos Cancer Institute (Detroit, MI), and Jonas de Souza, MD, MBA, assistant professor of medicine at The University of Chicago Medicine (Chicago, IL), evaluated the ASCO value framework within the scope of novel therapies developed for chronic lymphocytic leukemia (CLL).

To determine whether hematology RCTs adhered to the ASCO and ESMO frameworks, independent reviewers scored trial outcomes based on published guidelines. These included two versions of the ASCO framework (version 1 scoring range, -20 to 130; version 2, lower range undefined to 180) and the ESMO Magnitude of Clinical Benefit Scale (range, 1-5).

Dr Cheung and colleagues reviewed 23 RCTs, noting that seven studies used hematology-specific endpoints—including time-to-progression, cytogenetic response, and symptomatic response—that differed from overall survival and progression-free survival, the main endpoints used to design the ASCO and ESMO frameworks. The median ASCO framework scores were 24 (interquartile range [IQR], 22-40; minimum, 6; maximum, 53) by version 1 and 26.7 (IQR, 17.4-37.6; minimum, -33.3; maximum, 116.3) by version 3. The median ESMO score was 2 (IQR, 2-3; minimum, 1; maximum, 4).

Scoring disagreements occurred for 10 RCTs under version 1 of the ASCO framework and for 14 RCTs under version 2; scoring disagreements occurred for 12 RCTs using ESMO’s framework. Divergent interpretations of the scoring system served as the primary driver of disagreement for version 1 of the ASCO framework, whereas differences in scoring toxicities served as the most commonly cited reason for version 2. For the ESMO framework, divergent scoring interpretations or missing progression-free survival data served as the primary reasons for disagreement. Further, two studies could not be evaluated under version 1 of the ASCO model, and two studies could not be evaluated using the ESMO scale.

The researchers observed a correlation coefficient of 0.10 (95% CI, -0.37 to 0.53) between version 1 of the ASCO model and the ESMO model, and a coefficient of -0.21 (95% CI, -0.59 to 0.24) between version 2 and ESMO. The coefficient for version 1 of the ASCO model vs version 2 was 0.30 (95% CI, -0.15 to 0.65).

“When studies could be scored, the correlations between ASCO (version 1 and version 2) and ESMO results were poor, suggesting that these frameworks may not reliably identify the value of therapies in hematology,” wrote Dr Cheung and colleagues. “Consideration for the unique outcomes and toxicities in this population is warranted. The ASCO and ESMO frameworks continue to evolve, and a partnership with societies representing hematologists and their patients would be fruitful.”

Drs Seymour and de Souza compared the most current version of the ASCO framework to current National Comprehensive Cancer Network (NCCN) CLL guidelines to determine whether ASCO’s value models for efficacy, toxicity, and cost of therapy could be applied to novel CLL therapies. 

The study included data from 39 CLL RCTs, of which 20% (n = 8) could be assessed using the ASCO framework. The researchers found that they were able to assess clinical benefit in contrast to cost for certain therapies. “When comparing Imbruvica (ibrutinib; Pharmacyclics, Janssen) plus Treanda (bendamustine; Teva Oncology) and Rituxan (rituximab; Genentech/Biogen Idec) to Treanda and Rituxan alone, the framework is able to demonstrate a high clinical benefit of Imbruvica along with its high cost-sharing,” Dr Seymour and de Souza wrote. “It is also able to demonstrate that Lemtrada (alemtuzumab; Sanofi, Genzyme) has lower clinical benefit compared to other regimens studied against chlorambucil with a higher drug acquisition cost.”

The researchers found that toxicity scores varied by drug, and that net health benefit scores were affected by the clinical variable used. They noted that six evaluable studies included control arms that, although cheaper, were “among the lowest preferred NCCN recommendations.” They concluded that because an RCT directly comparing ibrutinib to chemoimmunotherapy is lacking, one would need to be performed in order to better understand the cost difference between the strategies.

“The ASCO Value Framework is a promising tool that helps compare costs and net health benefits,” Dr Seymour and de Souza wrote. “Major limitations when applied to CLL include limited number of trials that could be evaluated by the Framework, and lack of trials comparing stronger control arms. To optimize application of this Framework, we would urge investigators and sponsors to consider the assessment and reporting of the required variables to determine the net health benefit of therapies in clinical trials.” —Cameron Kelsall