A framework for understanding and targeting residual disease in oncogene-driven solid cancers

Abstract

Molecular targeted therapy has the potential to dramatically improve survival in patients with cancer. However, complete and durable responses to targeted therapy are rare in individuals with advanced-stage solid cancers. Even the most effective targeted therapies generally do not induce a complete tumor response, resulting in residual disease and tumor progression that limits patient survival. We discuss the emerging need to more fully understand the molecular basis of residual disease as a prelude to designing therapeutic strategies to minimize or eliminate residual disease so that we can move from temporary to chronic control of disease, or a cure, for patients with advanced-stage solid cancers. Ultimately, we propose a shift from the current reactive paradigm of analyzing and treating acquired drug resistance to a pre-emptive paradigm of defining the mechanisms that result in residual disease, to target and limit this disease reservoir.

Figure 1. The clinical problem of residual disease during targeted cancer therapy

A prototypical waterfall plot of best response to therapy highlighting the difference between intrinsic drug resistance and the residual tumor burden observed even amongst patients deemed to have an objective tumor response by RECIST criteria.

Figure 2. Modes of residual disease and therapy failure

a, Intrinsic resistance describes the survival of subpopulations of drug-resistant tumor cells within a generally sensitive tumor during initial therapy. Shown is a pre-treatment melanoma harboring different clones of cells, some with BRAF^V600E and some with BRAF^WT that instead have mutant PI3K. These resistant BRAF^WT cells form a drug-resistant niche during initial BRAFi treatment that results in incomplete response and eventual therapy failure and tumor progression. b, Tumor-cell adaptation can occur via therapy-induced changes in the tumor cells that enable adaptive survival and/or drug-tolerance, fueling a drug-resistant residual disease niche at maximal response. Here, the NF-kB-STAT3 axis, a JARID1A-mediated epigenetic program, and/or secreted factors including IL6 and HGF promote a drug-tolerant state that enables lung cancer cell persistence during initial EGFR inhibitor (EGFRi) treatment. c, Pharmacokinetic therapy failure can result from either pharmacologic limitations or dose-limiting toxicities or tumor intrinsic barriers to drug penetration into the tumor-cell compartment. Shown is poor penetration of a BRAF inhibitor (BRAFi) into a melanoma with dense stromal cell infiltration (red), resulting in low efficiency kill of BRAF^V600E-mutant tumor cells (green). The centered dotted circle (grey) indicates the functional overlap and continuum across these modes of residual disease driving therapy failure.

Figure 3. Multimodal characterization of residual disease and rational therapeutic targeting via upfront, iterative polytherapy

a, Multimodal longitudinal characterization of the molecular and cellular features of the tumor before therapy and at incomplete maximal response to reveal co-targets is necessary to understand the biologic underpinnings of residual disease (highlighted in the shaded box; measured and characterized as the residual tumor fraction, RTF). This analysis could be accomplished via serial sampling of tumor and liquid biopsies, which offer key complementary information regarding genetic clones, adaptive signaling events, and cellular constituents within the tumor microenvironment during the evolution of residual disease and therapy failure. Emerging single-cell genetic and proteomic analytic methods could be incorporated^,. Here, multiple modes of residual disease including, (1) therapy-induced selection of resistant clones and (2) induction of tumor-cell adaptive programs, contribute to the drug-resistant residual disease niche that enables eventual therapy-resistant tumor progression. b, Shown is a conceptual approach to rational upfront polytherapy that addresses the relevant tumor-cell clones present prior to therapy by co-inhibiting a critical downstream effector MEK together with a key on-target EGFR inhibitor resistance mutation (EGFR^T790M). Serial analysis of the residual state, smaller upon (b) polytherapy vs. (a) monotherapy, following initial polytherapy then identifies persistent tumor cells with adaptive STAT3 activation that is, in turn, treated with combined JAK inhibition together with an immune-cell activating therapy (here, anti-PD1). This results in further tumor-cell adaptation that is then controlled by the therapy-mediated activation of the host immune system. This strategy highlights the potential of rational upfront polytherapy, here using simultaneous and sequential combination regimens, to control and limit residual disease as well as toxicity, preventing overt tumor progression via continuous monitoring that guides iterative rational polytherapy. Arrows indicate tumor or liquid biopsy, as in (a). c, Shown are monotherapy versus rational polytherapy for ALK+ lung cancers, where either EGFR or MEK inhibition is paired with ALK inhibition to limit EGFR and MEK signaling that blunts ALK inhibitor response. Targeting upstream EGFR in ALK+ cancer cells may block MAPK and also PI3K/mTOR signaling among others, but may not be effective in patients, for example, with RAS gene amplification or mutation. Inhibiting MEK may more broadly target escape pathways in ALK+ cancer regardless of the upstream initiator of MEK activation (for example, EGFR or RAS), but may allow other escape pathways (for example, PI3K or NF-kB).

Figure 3. Multimodal characterization of residual disease and rational therapeutic targeting via upfront, iterative polytherapy

a, Multimodal longitudinal characterization of the molecular and cellular features of the tumor before therapy and at incomplete maximal response to reveal co-targets is necessary to understand the biologic underpinnings of residual disease (highlighted in the shaded box; measured and characterized as the residual tumor fraction, RTF). This analysis could be accomplished via serial sampling of tumor and liquid biopsies, which offer key complementary information regarding genetic clones, adaptive signaling events, and cellular constituents within the tumor microenvironment during the evolution of residual disease and therapy failure. Emerging single-cell genetic and proteomic analytic methods could be incorporated^,. Here, multiple modes of residual disease including, (1) therapy-induced selection of resistant clones and (2) induction of tumor-cell adaptive programs, contribute to the drug-resistant residual disease niche that enables eventual therapy-resistant tumor progression. b, Shown is a conceptual approach to rational upfront polytherapy that addresses the relevant tumor-cell clones present prior to therapy by co-inhibiting a critical downstream effector MEK together with a key on-target EGFR inhibitor resistance mutation (EGFR^T790M). Serial analysis of the residual state, smaller upon (b) polytherapy vs. (a) monotherapy, following initial polytherapy then identifies persistent tumor cells with adaptive STAT3 activation that is, in turn, treated with combined JAK inhibition together with an immune-cell activating therapy (here, anti-PD1). This results in further tumor-cell adaptation that is then controlled by the therapy-mediated activation of the host immune system. This strategy highlights the potential of rational upfront polytherapy, here using simultaneous and sequential combination regimens, to control and limit residual disease as well as toxicity, preventing overt tumor progression via continuous monitoring that guides iterative rational polytherapy. Arrows indicate tumor or liquid biopsy, as in (a). c, Shown are monotherapy versus rational polytherapy for ALK+ lung cancers, where either EGFR or MEK inhibition is paired with ALK inhibition to limit EGFR and MEK signaling that blunts ALK inhibitor response. Targeting upstream EGFR in ALK+ cancer cells may block MAPK and also PI3K/mTOR signaling among others, but may not be effective in patients, for example, with RAS gene amplification or mutation. Inhibiting MEK may more broadly target escape pathways in ALK+ cancer regardless of the upstream initiator of MEK activation (for example, EGFR or RAS), but may allow other escape pathways (for example, PI3K or NF-kB).