Cycl2r, you have identified an important tension in the CLOVER trial, and I'll answer it as completely as I can.
The short version: your concern is real, your logic is correct, and the fourth activity layer is directly relevant to resolving it, but not in isolation. The complete answer requires understanding what the TAS-102 backbone alone produces at RECIST, and what leronlimab's chemotherapy sensitization mechanism does to that number specifically.
The longer version follows.
The primary endpoint of CLOVER is objective response rate as defined by RECIST v1.1 criteria, in CCR5-positive, MSS, relapsed and refractory mCRC patients receiving leronlimab in combination with TAS-102 plus bevacizumab. You are correct that RECIST v1.1 ORR requires at least a 30 percent reduction in the sum of target lesion diameters to qualify as a partial response, and a complete disappearance to qualify as a complete response.
ctDNA clearance, however dramatic, does not satisfy that criterion. The FDA adjudicates on RECIST. That is the published protocol and it does not flex for molecular response data alone. https://www.onclive.com/view/fda-approves-sac...ast-cancer
So the question you are really asking is: what does the backbone produce at RECIST without leronlimab, and what does leronlimab's fourth activity layer contribute to that number? About 2 years ago, I wrote about ORR: https://www.reddit.com/r/Livimmune/comments/1...e_changer/
The SUNLIGHT phase 3 trial answers the backbone question with precision. The ORR of patients treated with bevacizumab plus TAS-102 was 6.3 percent versus 0.9 percent for TAS-102 alone, with a disease control rate of 76.6 percent for the combination versus 47 percent for TAS-102 alone. This is the published benchmark for the standard of care backbone in this population.
6.3 percent of patients on TAS-102 plus bevacizumab alone achieve RECIST-defined tumor shrinkage of at least 30 percent. The bar is low precisely because this population is heavily pretreated and the backbone's primary value is disease stabilization and overall survival extension, not tumor shrinkage. https://www.sciencedirect.com/science/article...2822000877
This is where the fourth activity layer becomes directly relevant to your question, and not as a rescue but as the primary driver of RECIST response above the backbone baseline.
TAS-102 works by incorporating trifluridine into tumor cell DNA, causing DNA strand breaks and replication errors. Bevacizumab limits the tumor's ability to recruit new blood vessels to repair and sustain itself. Both mechanisms ultimately depend on the tumor cell's DNA damage response machinery to determine whether the cell dies or repairs and survives.
The CCR5-dependent proliferative signaling cascade, specifically the CEP131, KHDRBS1, and MAPK6 cell cycle regulators whose activation is completely abolished by CCR5 knockout in the June 2026 Frontiers in Immunology phosphoproteomics paper, governs the tumor cell's ability to execute that repair response. When leronlimab blocks CCR5 simultaneously with TAS-102 delivering DNA damage, the tumor cell is facing strand breaks while its repair machinery is metabolically and biochemically compromised at the cell cycle level. The Seahorse metabolic profiling from the AACR Brain Cancer poster documented this directly: CCR5 inhibition reduced oxygen consumption rate in a dose-dependent manner, meaning the tumor cell's metabolic capacity to fuel the repair response is reduced at the same time the damage is being delivered.
This is the mechanical basis for expecting CLOVER's RECIST ORR to substantially exceed the 6.3 percent SUNLIGHT benchmark. Not because leronlimab independently kills tumor cells, but because it removes the metabolic and cell cycle infrastructure the tumor cell requires to survive TAS-102's DNA damage. More cells that would have repaired and survived the backbone chemotherapy instead die and meet RECIST response criteria.
The ctDNA picture adds a third layer of interpretive context. A median 70 percent ctDNA decline across evaluable CLOVER patients at week two, with four reaching completely undetectable levels, documents that tumor cell proliferative activity is being suppressed at the molecular level before the eight-week RECIST scan. Tumors in which circulating tumor DNA is declining sharply are tumors in which the replication machinery is being disrupted. Whether that disruption translates into enough cell death to produce 30 percent volumetric shrinkage measurable by CT imaging depends on the rate of cell death relative to the rate of residual stromal and fibrotic tissue contraction within the tumor mass. These are not always the same number, which is why ctDNA and RECIST can diverge. But a tumor in which ctDNA is collapsing at week two is a tumor in which the fourth activity layer is operating exactly as the framework would predict.
The confirmed ORR at ESMO in October will tell us how many patients crossed the 30 percent RECIST threshold. The molecular data says the mechanism is active. The RECIST scan says whether enough cells died fast enough to show up on imaging. Both questions have the same answer waiting in the same dataset. October is the moment both become visible simultaneously.
Your concern was precisely framed. The fourth activity layer is the direct mechanical answer to it, and the SUNLIGHT 6.3 percent baseline is the number it needs to exceed to demonstrate that leronlimab is doing something RECIST-meaningful to the backbone's cytotoxic capacity.
The short version: your concern is real, your logic is correct, and the fourth activity layer is directly relevant to resolving it, but not in isolation. The complete answer requires understanding what the TAS-102 backbone alone produces at RECIST, and what leronlimab's chemotherapy sensitization mechanism does to that number specifically.
The longer version follows.
The primary endpoint of CLOVER is objective response rate as defined by RECIST v1.1 criteria, in CCR5-positive, MSS, relapsed and refractory mCRC patients receiving leronlimab in combination with TAS-102 plus bevacizumab. You are correct that RECIST v1.1 ORR requires at least a 30 percent reduction in the sum of target lesion diameters to qualify as a partial response, and a complete disappearance to qualify as a complete response.
ctDNA clearance, however dramatic, does not satisfy that criterion. The FDA adjudicates on RECIST. That is the published protocol and it does not flex for molecular response data alone. https://www.onclive.com/view/fda-approves-sac...ast-cancer
So the question you are really asking is: what does the backbone produce at RECIST without leronlimab, and what does leronlimab's fourth activity layer contribute to that number? About 2 years ago, I wrote about ORR: https://www.reddit.com/r/Livimmune/comments/1...e_changer/
The SUNLIGHT phase 3 trial answers the backbone question with precision. The ORR of patients treated with bevacizumab plus TAS-102 was 6.3 percent versus 0.9 percent for TAS-102 alone, with a disease control rate of 76.6 percent for the combination versus 47 percent for TAS-102 alone. This is the published benchmark for the standard of care backbone in this population.
6.3 percent of patients on TAS-102 plus bevacizumab alone achieve RECIST-defined tumor shrinkage of at least 30 percent. The bar is low precisely because this population is heavily pretreated and the backbone's primary value is disease stabilization and overall survival extension, not tumor shrinkage. https://www.sciencedirect.com/science/article...2822000877
This is where the fourth activity layer becomes directly relevant to your question, and not as a rescue but as the primary driver of RECIST response above the backbone baseline.
TAS-102 works by incorporating trifluridine into tumor cell DNA, causing DNA strand breaks and replication errors. Bevacizumab limits the tumor's ability to recruit new blood vessels to repair and sustain itself. Both mechanisms ultimately depend on the tumor cell's DNA damage response machinery to determine whether the cell dies or repairs and survives.
The CCR5-dependent proliferative signaling cascade, specifically the CEP131, KHDRBS1, and MAPK6 cell cycle regulators whose activation is completely abolished by CCR5 knockout in the June 2026 Frontiers in Immunology phosphoproteomics paper, governs the tumor cell's ability to execute that repair response. When leronlimab blocks CCR5 simultaneously with TAS-102 delivering DNA damage, the tumor cell is facing strand breaks while its repair machinery is metabolically and biochemically compromised at the cell cycle level. The Seahorse metabolic profiling from the AACR Brain Cancer poster documented this directly: CCR5 inhibition reduced oxygen consumption rate in a dose-dependent manner, meaning the tumor cell's metabolic capacity to fuel the repair response is reduced at the same time the damage is being delivered.
This is the mechanical basis for expecting CLOVER's RECIST ORR to substantially exceed the 6.3 percent SUNLIGHT benchmark. Not because leronlimab independently kills tumor cells, but because it removes the metabolic and cell cycle infrastructure the tumor cell requires to survive TAS-102's DNA damage. More cells that would have repaired and survived the backbone chemotherapy instead die and meet RECIST response criteria.
The ctDNA picture adds a third layer of interpretive context. A median 70 percent ctDNA decline across evaluable CLOVER patients at week two, with four reaching completely undetectable levels, documents that tumor cell proliferative activity is being suppressed at the molecular level before the eight-week RECIST scan. Tumors in which circulating tumor DNA is declining sharply are tumors in which the replication machinery is being disrupted. Whether that disruption translates into enough cell death to produce 30 percent volumetric shrinkage measurable by CT imaging depends on the rate of cell death relative to the rate of residual stromal and fibrotic tissue contraction within the tumor mass. These are not always the same number, which is why ctDNA and RECIST can diverge. But a tumor in which ctDNA is collapsing at week two is a tumor in which the fourth activity layer is operating exactly as the framework would predict.
The confirmed ORR at ESMO in October will tell us how many patients crossed the 30 percent RECIST threshold. The molecular data says the mechanism is active. The RECIST scan says whether enough cells died fast enough to show up on imaging. Both questions have the same answer waiting in the same dataset. October is the moment both become visible simultaneously.
Your concern was precisely framed. The fourth activity layer is the direct mechanical answer to it, and the SUNLIGHT 6.3 percent baseline is the number it needs to exceed to demonstrate that leronlimab is doing something RECIST-meaningful to the backbone's cytotoxic capacity.
8
