In:
Cancer Research, American Association for Cancer Research (AACR), Vol. 79, No. 13_Supplement ( 2019-07-01), p. SY45-03-SY45-03
Abstract:
Cancer heterogeneity is a major hurdle for the development of efficient therapeutic strategies. Patient selection strategies that include very specific molecular markers are needed to reliably obtain response during clinical development and ultimately bring new therapeutics to patients. Heterogeneity is defined at multiple levels of granularity: Across patients, across lesions in a patient, across areas of a lesion and across cells within a lesion area. At the simplest level, some patient to patient major genetic variations are well established as predictors of drug response (such as BRAF mutation or EGFR mutation). However, even within genetically defined responsive groups there is often a broad range of clinical responses. This heterogenous response can sometime be attributed to additional genetic differences found in poorly responsive patients. In some cases, however no clear genetic pattern emerges to explain clinical variations. In the context of acquired resistance there are also clear examples of secondary mutations in initially sensitive patients that can explain resistance. Nevertheless, this is again not always the case and there are clear examples of acquired resistance that are not accompanied by genetic changes. In some of these cases, however drug combinations that allow to restore sensitivity can be identified through systematic drug and genetic screens. There is thus an important phenotypic heterogeneity that is not well defined when considering only genetic information. In addition, how diverse mutational pattern across genes result in diverse of consistent drug response phenotypes is poorly understood. Consequently, the genetic and phenotypic heterogeneity are overall not mechanistically well linked to each other. In fact, whether or not phenotypic heterogeneity is less or more than the heterogeneity estimated through genetic annotation is unclear. A better understanding of biological pathways and their interplay is still required to organize the genetic information in meaningful functional groups. Systematic testing of candidate strategies through genetic and drug screening across laboratory models that are representative of the disease can provide insights into drug response heterogeneity and its molecular (genetic and otherwise) basis. A large number of models is needed to capture heterogeneity even within major genotypes to provide insight into these important therapeutic discovery issues. A specific aspect of how heterogeneity impacts therapeutic discovery is the following: Often time, the discovery made in a small numbers of laboratory models does not apply to the majority of tumors that the models were chosen to represent. The issue goes beyond addressing different oncogenic drivers and is more problematic when considering a given genetically define group (for example KRAS mutant non-small cell lung cancers). It is also not necessarily due to the lack of robustness of the findings in the initial model(s) but rather because these findings do not necessarily apply to other models (and clinical cases) broadly. This lack of “scalability” of the findings in discovery approaches precludes some otherwise well controlled but underpowered studies translate into actionable clinical strategies because they fail to capture the clinical heterogeneity. One approach to address the issue, is to use large collection of laboratory models such as tumor derived cell lines and short-term cultures or other tumor derived models in large numbers. Large enough collections allow to define, at least to some extent, the prevalence of the mechanisms identified, and the specificity of activity in the genetically defined group of patients that they seek to address. Tumor heterogeneity is a particularly acute challenge in the area of drug combination discovery. Combining drugs is largely considered to be a necessary path to clinical success, particularly in solid tumors. Indeed, there is a large compendium of empirically discovered drug combination currently used in oncology. Rational design of combination should be more efficient and hopefully come with less burden on patients through lowering of toxicity. In principle, combination therapies can address patient heterogeneity by improving population coverage, by addressing different population of cancer cells within one patient or by affecting a given cell population more efficiently than single agents. The paradigm of synthetic lethality as well as principle of complete signal extinction resulting in more than additive outcome aim at addressing the later. Because the principles of drug response to a single drug are poorly defined and in particular often not always well explained by genetic information, the discovery of drug combinations based on single agent response profile and/or genetic information is inefficient. Systematic drug combinations screening is thus an important approach. It is however a challenging one because of the number of tests that need to be performed. Importantly, again in principle, drug combinations could address the heterogeneity of single agent responses. For instance, in cases where differential response within a genetically defined patient population is in majority due to a common mechanism that can be targeted using a second drug. For example, EGFR signaling limits the activity of BRAF_MEK combination in a substantial fraction of BRAF V600E mutant colorectal cancers and the triple combination has demonstrated benefit over the BRAF_MEK inhibitors combination. In many instances however, the issue at hand is more complex, there is very limited efficacy of single agent in any patient and combinations are sought out to obtain clinically meaningful responses. In these cases, open ended combinatorial exploratory studies need to be performed. For the results to be translatable, new combinations need to address either the majority of genetically defined patients or be accompanied by robust predictive biomarkers. I will present data and analyses towards addressing these diverse challenges in therapeutic discovery and application. I will discuss our findings on acquired therapeutic resistance in non-small cell lung cancer as well as our unpublished results on the systematic evaluation of drug combination in non-small cell lung cancer models. The results that will be discussed come from systematic efforts of drug screening in historically established cell lines as well as newly derived cell lines from clinical biopsies. I will address how studies on large cancer model collection inform the results of focused deep screening seeking to identify new targets and synthetic lethal activities. Citation Format: Cyril Benes, Cyril H. Benes. Capturing the therapeutic response heterogeneity at the functional level [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr SY45-03.
Type of Medium:
Online Resource
ISSN:
0008-5472
,
1538-7445
DOI:
10.1158/1538-7445.AM2019-SY45-03
Language:
English
Publisher:
American Association for Cancer Research (AACR)
Publication Date:
2019
detail.hit.zdb_id:
2036785-5
detail.hit.zdb_id:
1432-1
detail.hit.zdb_id:
410466-3
Bookmarklink
