Adrie Van Bokhoven
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Adrie Van Bokhoven has a total of 176 co-authors. Here are the top co-authors by number of shared publications:
- Alyse W. Staley
University of Colorado Anschutz Medical Campus 2 Department of Pediatrics Aurora, Colorado
32 shared publications - Andrew Goodspeed
University of Colorado Cancer Center, University of Colorado Anschutz Medical Campus, Aurora, Colorado
32 shared publications - David J. Orlicky
University of Colorado Anschutz School of Medicine
32 shared publications - Jennifer A. Wagner
University of Colorado Anschutz Medical Campus 1 Department of Pathology Aurora, Colorado
32 shared publications - Jennifer K. Richer
University of Colorado Anschutz Medical Campus 1 Department of Pathology Aurora, Colorado
32 shared publications - Jennifer R. Diamond
University of Colorado Anschutz Medical Campus 5 Division of Medical Oncology Aurora, Colorado
32 shared publications
Latest Publications
- Open Access
Tumor Associated Neutrophils Extend NETs To Spatially Drive Necroptosis In Glioblastoma
Background: Glioblastoma is pathologically characterized by extensive pseudopalisading necrosis, a hallmark linked to therapeutic resistance and dismal prognosis. While traditionally viewed as a passive ischemic event, the precise radiogenomic drivers and spatial immune architectures orchestrating active tumor necrosis remain poorly understood. Methods: We integrated macroscopic clinical MRI radiomics with public genomic datasets (n=3,639), single-cell spatial transcriptomics, and high-resolution protein immunofluorescence (IF) assays in a novel co-mutant murine model to map the necrotic microenvironment. Results: Radiomic analysis of patient MRIs demonstrated that high tumoral necrotic volume serves as a powerful independent biomarker for poor long-term survival. This macro-structural necrosis phenotype is radiogenomically driven by cell-intrinsic TP53 , PTEN , and NF1 (PPN) co-mutations. Spatial transcriptomic profiling of PPN-deficient tumors revealed distinct cellular topography, separating a proliferative, invasive niche (EGFR + SOX + , Type Inv ) from a hyper-aggressive necrotic niche (ERBB3 + , Type Nec ). Mechanistically, Type Nec cells deploy a prominent CCL5/CXCL2 chemokine milieu that actively recruits and prompts tumor-associated neutrophils to undergo neutrophil extracellular trap (NET) formation. This overactive neutrophil infiltrate drives a feed-forward loop of intrinsic tumor necroptosis mediated by the release of the damage-associated molecular pattern (DAMP) protein HMGB2. This cascade could be therapeutically dismantled using targeted anti-inflammatory compounds. Finally, multi-cohort clinical validation confirmed that an HMGB2 high signature dictates a high-risk, myeloid-rich immunosuppressive microenvironment in glioblastoma patients. Conclusions: Our findings redefine glioblastoma necrosis as an active, mutation-driven immunopathological process. The HMGB2-mediated DAMP necroptosis cascade and associated myeloid chemokines provide robust prognostic biomarkers and immediate therapeutic vulnerabilities to disrupt the immunosuppressive necrotic niche. Key Points: Clinical Stratification via Radiomics: Preoperative MRI analysis of 1,079 glioblastoma patients establishes a necrotic core volume threshold of 18.57% as a robust, non-invasive biomarker for early survival divergence and dismal long-term prognosis. Genomic Orchestration of Necrosis: Cross-species modeling identifies a high-risk somatic PPN co-mutation profile yielding two topographically distinct niches: an invasive niche EGFR + SOX + , Type Inv ) from a hyper-aggressive necrotic niche (ERBB3 + , Type Nec ). Vascular Loop and DAMP Cascade: TypeNec cells deploy a CCL5/CXCL2 chemokine milieu driving perivascular neutrophil recruitment and MPO + CitH3 + NETosis, sparking a feed-forward loop of hypoxia and necrosis driven by tumor-derived HMGB2 high that is pharmacologically targetable. Online Key Points (85 characters): Preoperative MRI biomarker of necrotic threshold determines long-term prognosis. Cross-species co-mutation profile yielding distinct invasive and necrotic niches. N eutrophil extracellular traps spark DAMP-targetable feed-forward necroptosis loop. Importance of the Study: Tumor necrosis is a hallmark diagnostic feature of glioblastoma, historically characterized as a passive consequence of local tissue ischemia. This study redefines glioblastoma necrosis as an active, genetically driven immunopathological process. By integrating macroscopic clinical MRI radiomics from over 1,000 patients with single-cell spatial transcriptomics and a novel co-mutant mouse model, we reveal that cell-intrinsic TP53 , PTEN , and NF1 co-mutations actively dictate a hyper-aggressive necrotic niche. Mechanistically, these mutated tumor cells deploy a CCL5/CXCL2 chemokine cascade to recruit neutrophils, triggering an overactive neutrophil extracellular trap (NET) and a HMGB2-mediated damage-associated molecular pattern (DAMP) loop that accelerates tissue destruction and myeloid immunosuppression. Crucially, we demonstrate that this pathogenic axis can be tracked non-invasively via standard neuroimaging and pharmacologically dismantled with anti-inflammatory small molecules, providing immediate diagnostic stratifications and a brand-new therapeutic vulnerability to disrupt the glioblastoma microenvironment. Lay Summary: Dead tissue zones inside glioblastoma brain tumors, known as necrosis, are tied to poor patient survival. While long believed to be an accidental result of rapid tumor growth cutting off oxygen, this study reveals that tumor cells actively trigger this cell death. Driven by specific genetic mutations, these cells release distress signals that recruit immune cells called neutrophils. These neutrophils create a destructive, inflammatory loop that worsens the disease. By combining patient MRI scans with advanced spatial tissue mapping, we show that this destructive process can be tracked non-invasively and stopped using anti-inflammatory drugs, offering a new treatment strategy.
Open AccessTumor Associated Neutrophils Extend NETs To Spatially Drive Necroptosis In Glioblastoma
Background: Glioblastoma is pathologically characterized by extensive pseudopalisading necrosis, a hallmark linked to therapeutic resistance and dismal prognosis. While traditionally viewed as a passive ischemic event, the precise radiogenomic drivers and spatial immune architectures orchestrating active tumor necrosis remain poorly understood. Methods: We integrated macroscopic clinical MRI radiomics with public genomic datasets (n=3,639), single-cell spatial transcriptomics, and high-resolution protein immunofluorescence (IF) assays in a novel co-mutant murine model to map the necrotic microenvironment. Results: Radiomic analysis of patient MRIs demonstrated that high tumoral necrotic volume serves as a powerful independent biomarker for poor long-term survival. This macro-structural necrosis phenotype is radiogenomically driven by cell-intrinsic TP53 , PTEN , and NF1 (PPN) co-mutations. Spatial transcriptomic profiling of PPN-deficient tumors revealed distinct cellular topography, separating a proliferative, invasive niche (EGFR + SOX + , Type Inv ) from a hyper-aggressive necrotic niche (ERBB3 + , Type Nec ). Mechanistically, Type Nec cells deploy a prominent CCL5/CXCL2 chemokine milieu that actively recruits and prompts tumor-associated neutrophils to undergo neutrophil extracellular trap (NET) formation. This overactive neutrophil infiltrate drives a feed-forward loop of intrinsic tumor necroptosis mediated by the release of the damage-associated molecular pattern (DAMP) protein HMGB2. This cascade could be therapeutically dismantled using targeted anti-inflammatory compounds. Finally, multi-cohort clinical validation confirmed that an HMGB2 high signature dictates a high-risk, myeloid-rich immunosuppressive microenvironment in glioblastoma patients. Conclusions: Our findings redefine glioblastoma necrosis as an active, mutation-driven immunopathological process. The HMGB2-mediated DAMP necroptosis cascade and associated myeloid chemokines provide robust prognostic biomarkers and immediate therapeutic vulnerabilities to disrupt the immunosuppressive necrotic niche. Key Points: Clinical Stratification via Radiomics: Preoperative MRI analysis of 1,079 glioblastoma patients establishes a necrotic core volume threshold of 18.57% as a robust, non-invasive biomarker for early survival divergence and dismal long-term prognosis. Genomic Orchestration of Necrosis: Cross-species modeling identifies a high-risk somatic PPN co-mutation profile yielding two topographically distinct niches: an invasive niche EGFR + SOX + , Type Inv ) from a hyper-aggressive necrotic niche (ERBB3 + , Type Nec ). Vascular Loop and DAMP Cascade: TypeNec cells deploy a CCL5/CXCL2 chemokine milieu driving perivascular neutrophil recruitment and MPO + CitH3 + NETosis, sparking a feed-forward loop of hypoxia and necrosis driven by tumor-derived HMGB2 high that is pharmacologically targetable. Online Key Points (85 characters): Preoperative MRI biomarker of necrotic threshold determines long-term prognosis. Cross-species co-mutation profile yielding distinct invasive and necrotic niches. N eutrophil extracellular traps spark DAMP-targetable feed-forward necroptosis loop. Importance of the Study: Tumor necrosis is a hallmark diagnostic feature of glioblastoma, historically characterized as a passive consequence of local tissue ischemia. This study redefines glioblastoma necrosis as an active, genetically driven immunopathological process. By integrating macroscopic clinical MRI radiomics from over 1,000 patients with single-cell spatial transcriptomics and a novel co-mutant mouse model, we reveal that cell-intrinsic TP53 , PTEN , and NF1 co-mutations actively dictate a hyper-aggressive necrotic niche. Mechanistically, these mutated tumor cells deploy a CCL5/CXCL2 chemokine cascade to recruit neutrophils, triggering an overactive neutrophil extracellular trap (NET) and a HMGB2-mediated damage-associated molecular pattern (DAMP) loop that accelerates tissue destruction and myeloid immunosuppression. Crucially, we demonstrate that this pathogenic axis can be tracked non-invasively via standard neuroimaging and pharmacologically dismantled with anti-inflammatory small molecules, providing immediate diagnostic stratifications and a brand-new therapeutic vulnerability to disrupt the glioblastoma microenvironment. Lay Summary: Dead tissue zones inside glioblastoma brain tumors, known as necrosis, are tied to poor patient survival. While long believed to be an accidental result of rapid tumor growth cutting off oxygen, this study reveals that tumor cells actively trigger this cell death. Driven by specific genetic mutations, these cells release distress signals that recruit immune cells called neutrophils. These neutrophils create a destructive, inflammatory loop that worsens the disease. By combining patient MRI scans with advanced spatial tissue mapping, we show that this destructive process can be tracked non-invasively and stopped using anti-inflammatory drugs, offering a new treatment strategy.
- Open Access
- 8 October 2026
- Published by American Association for Cancer Research (AACR)
Data from Dissociated Response to Immune Checkpoint Inhibition in Head and Neck Squamous Cell Carcinoma Is Common and Exhibits Distinct Clinical Outcomes
Immune checkpoint inhibition (ICI) constitutes the first-line therapy for relapsed/metastatic (R/M) head and neck squamous cell carcinoma (HNSCC) despite low response rates. Some patients treated with ICI experience dissociated response (DR), the coexistence of responding and progression lesions. DR prevalence and prognostic impact in HNSCC are unknown. We retrospectively identified 112 patients treated with ICI at our institution and evaluated their disease characteristics, survival outcomes, and response. Tissue specimens were used to assess the tumor microenvironment (TME). DR occurred in 19% of our patient cohort. DR was more likely in patients with disease involving the cervical lymph nodes [odds ratio (OR), 16.79; P = 0.004]. Patients with DR had a median progression-free and overall survival of 6.7 and 16.9 months, respectively, similar to subjects with partial response (PR) or stable disease (SD). DR was less favorable when it occurred early, and we identified no instances in which early DR converted to uniform objective response. We demonstrated increased CD8+ T-cell infiltration and decreased signal transducer and activator of transcription 3 (STAT3) signaling in tissue from responding versus nonresponding patients. Tumors from patients with DR had low T-cell infiltration and reduced STAT3. In conclusion, DR during ICI therapy in HNSCC is common and associated with favorable prognosis, especially in patients who initially have a complete response (CR) or PR. Early DR as the best response is less beneficial and unlikely to convert to a true response; change in therapy should be considered for these patients. Response patterns may be related to organ-specific differences in the TME. Significance:Mixed response or DR is common in patients with R/M HNSCC who receive immune checkpoint inhibition (ICI). Patients with DR have survival outcomes similar to patients with a PR, especially when DR occurs after an initial uniform response to therapy. DR occurring early in treatment is less favorable and unlikely to convert to uniform response later on.
- 8 October 2026
- Published by American Association for Cancer Research (AACR)
Open AccessData from Dissociated Response to Immune Checkpoint Inhibition in Head and Neck Squamous Cell Carcinoma Is Common and Exhibits Distinct Clinical Outcomes
Immune checkpoint inhibition (ICI) constitutes the first-line therapy for relapsed/metastatic (R/M) head and neck squamous cell carcinoma (HNSCC) despite low response rates. Some patients treated with ICI experience dissociated response (DR), the coexistence of responding and progression lesions. DR prevalence and prognostic impact in HNSCC are unknown. We retrospectively identified 112 patients treated with ICI at our institution and evaluated their disease characteristics, survival outcomes, and response. Tissue specimens were used to assess the tumor microenvironment (TME). DR occurred in 19% of our patient cohort. DR was more likely in patients with disease involving the cervical lymph nodes [odds ratio (OR), 16.79; P = 0.004]. Patients with DR had a median progression-free and overall survival of 6.7 and 16.9 months, respectively, similar to subjects with partial response (PR) or stable disease (SD). DR was less favorable when it occurred early, and we identified no instances in which early DR converted to uniform objective response. We demonstrated increased CD8+ T-cell infiltration and decreased signal transducer and activator of transcription 3 (STAT3) signaling in tissue from responding versus nonresponding patients. Tumors from patients with DR had low T-cell infiltration and reduced STAT3. In conclusion, DR during ICI therapy in HNSCC is common and associated with favorable prognosis, especially in patients who initially have a complete response (CR) or PR. Early DR as the best response is less beneficial and unlikely to convert to a true response; change in therapy should be considered for these patients. Response patterns may be related to organ-specific differences in the TME. Significance:Mixed response or DR is common in patients with R/M HNSCC who receive immune checkpoint inhibition (ICI). Patients with DR have survival outcomes similar to patients with a PR, especially when DR occurs after an initial uniform response to therapy. DR occurring early in treatment is less favorable and unlikely to convert to uniform response later on.
- Open Access
- 8 October 2026
- Published by American Association for Cancer Research (AACR)
Figure 1 from Dissociated Response to Immune Checkpoint Inhibition in Head and Neck Squamous Cell Carcinoma Is Common and Exhibits Distinct Clinical Outcomes
Patient selection process and DR examples. A, Initial screen identified patients with HNSCC treated with ICI. Patients were excluded if treated outside of the R/M setting, stopped ICI prior to receiving three cycles, or clinical records and/or imaging were incomplete. B, Radiology example of late DR. Patient had a large neck mass on baseline scans. Best response occurred at 6 months on treatment with substantial PR in the neck. Ten months after treatment start, the patient had a late DR with sustained PR in the neck but PD in the axillary lymph nodes which were treated locally with radiation. Overall PD occurred after 14 months on treatment with progression in the neck and diffusely through the lungs. C, Radiology example of early DR. Patient had a pleural-based mass on pretreatment scans. Best response and early DR occurred at 3 months on treatment with PR of the pleural-based lesion but PD in the right hilum. Patient continued on ICI alone and developed overall PD after 4 months on treatment with continued progression of hilar lymph nodes along with increased subcarinal and mediastinal lymphadenopathy.
- 8 October 2026
- Published by American Association for Cancer Research (AACR)
Open AccessFigure 1 from Dissociated Response to Immune Checkpoint Inhibition in Head and Neck Squamous Cell Carcinoma Is Common and Exhibits Distinct Clinical Outcomes
Patient selection process and DR examples. A, Initial screen identified patients with HNSCC treated with ICI. Patients were excluded if treated outside of the R/M setting, stopped ICI prior to receiving three cycles, or clinical records and/or imaging were incomplete. B, Radiology example of late DR. Patient had a large neck mass on baseline scans. Best response occurred at 6 months on treatment with substantial PR in the neck. Ten months after treatment start, the patient had a late DR with sustained PR in the neck but PD in the axillary lymph nodes which were treated locally with radiation. Overall PD occurred after 14 months on treatment with progression in the neck and diffusely through the lungs. C, Radiology example of early DR. Patient had a pleural-based mass on pretreatment scans. Best response and early DR occurred at 3 months on treatment with PR of the pleural-based lesion but PD in the right hilum. Patient continued on ICI alone and developed overall PD after 4 months on treatment with continued progression of hilar lymph nodes along with increased subcarinal and mediastinal lymphadenopathy.
- Open Access
- 8 October 2026
- Published by American Association for Cancer Research (AACR)
Supplemental Table 1 from Dissociated Response to Immune Checkpoint Inhibition in Head and Neck Squamous Cell Carcinoma Is Common and Exhibits Distinct Clinical Outcomes
Supplemental Table 1. Clinical characteristics associated with timing of DR.
- 8 October 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplemental Table 1 from Dissociated Response to Immune Checkpoint Inhibition in Head and Neck Squamous Cell Carcinoma Is Common and Exhibits Distinct Clinical Outcomes
Supplemental Table 1. Clinical characteristics associated with timing of DR.
- Open Access
- 8 October 2026
- Published by American Association for Cancer Research (AACR)
Figure 2 from Dissociated Response to Immune Checkpoint Inhibition in Head and Neck Squamous Cell Carcinoma Is Common and Exhibits Distinct Clinical Outcomes
Organ-specific treatment response rates and impact of overall treatment response on survival outcomes. A, Organ-specific treatment responses for local disease (n = 54), cervical lymphadenopathy (n = 42), lung metastases (n = 59), other (noncervical) lymph nodes (LN; n = 45), and other sites of disease (bone, liver, skin and soft tissue, and/or brain, n = 26). B, Kaplan–Meier estimates of PFS (left) and OS (right) in months. Top, RECIST response classifications. Middle, subdivided response categories demonstrated that DR was similar to SD and PR with respect to clinical outcomes. Bottom, late DR had more favorable impact on survival than early DR. There were 86 PFS events and 61 OS events in the total cohort, and 19 PFS events and 14 OS events in the DR subgroup. Tick marks indicate censored data.
- 8 October 2026
- Published by American Association for Cancer Research (AACR)
Open AccessFigure 2 from Dissociated Response to Immune Checkpoint Inhibition in Head and Neck Squamous Cell Carcinoma Is Common and Exhibits Distinct Clinical Outcomes
Organ-specific treatment response rates and impact of overall treatment response on survival outcomes. A, Organ-specific treatment responses for local disease (n = 54), cervical lymphadenopathy (n = 42), lung metastases (n = 59), other (noncervical) lymph nodes (LN; n = 45), and other sites of disease (bone, liver, skin and soft tissue, and/or brain, n = 26). B, Kaplan–Meier estimates of PFS (left) and OS (right) in months. Top, RECIST response classifications. Middle, subdivided response categories demonstrated that DR was similar to SD and PR with respect to clinical outcomes. Bottom, late DR had more favorable impact on survival than early DR. There were 86 PFS events and 61 OS events in the total cohort, and 19 PFS events and 14 OS events in the DR subgroup. Tick marks indicate censored data.
- Open Access
- 8 October 2026
- Published by American Association for Cancer Research (AACR)
Figure 3 from Dissociated Response to Immune Checkpoint Inhibition in Head and Neck Squamous Cell Carcinoma Is Common and Exhibits Distinct Clinical Outcomes
TILs are associated with treatment response patterns. A, Quantification and (B) IHC images of CD8+ T cells showed increased infiltration in the parenchyma of responders (R) vs. nonresponders (NR) or dissociated responders. No differences were observed in the stroma. Brown, CD8; red, CD3. C, Early DR cases demonstrated higher CD8+ T-cell invasion compared with late DR cases in both the parenchyma and stroma. Five representative images per case were quantified; lines represent mean, and errors bars represent standard error of the mean. DR, dissociated response.
- 8 October 2026
- Published by American Association for Cancer Research (AACR)
Open AccessFigure 3 from Dissociated Response to Immune Checkpoint Inhibition in Head and Neck Squamous Cell Carcinoma Is Common and Exhibits Distinct Clinical Outcomes
TILs are associated with treatment response patterns. A, Quantification and (B) IHC images of CD8+ T cells showed increased infiltration in the parenchyma of responders (R) vs. nonresponders (NR) or dissociated responders. No differences were observed in the stroma. Brown, CD8; red, CD3. C, Early DR cases demonstrated higher CD8+ T-cell invasion compared with late DR cases in both the parenchyma and stroma. Five representative images per case were quantified; lines represent mean, and errors bars represent standard error of the mean. DR, dissociated response.
- Open Access
- 8 October 2026
- Published by American Association for Cancer Research (AACR)
Figure 4 from Dissociated Response to Immune Checkpoint Inhibition in Head and Neck Squamous Cell Carcinoma Is Common and Exhibits Distinct Clinical Outcomes
Differential activity of PD-L1 interactome signaling based on tumor response. A, Quantification of PD-L1, ILF3, STAT3, and pSTAT3 showed increased ILF3 and STAT3 expression in nonresponding tumors. NR, nonresponders; R, responders. B, Representative IHC images of ILF3 and STAT3 staining by response type. C, All components of the PD-L1 interactome were increased in early compared with late DR cases. Five representative images per case were quantified; lines represent mean, and errors bars represent standard error of the mean. D, Top, differences in STAT3 expression by treatment response were specific to tissue samples collected just prior to ICI initiation. E, Multicolor PLA representative images (left) and quantification (right) revealed significantly higher PD-L1 intracellular signaling activity in nonresponding vs. responding cases as demonstrated by increased interaction between PD-L1–ILF3 and PD-L1–STAT3. Five cases per group with quantification from five representative images. Scale bars, 10 μm (left); magnified PLA is 4× original size.
- 8 October 2026
- Published by American Association for Cancer Research (AACR)
Open AccessFigure 4 from Dissociated Response to Immune Checkpoint Inhibition in Head and Neck Squamous Cell Carcinoma Is Common and Exhibits Distinct Clinical Outcomes
Differential activity of PD-L1 interactome signaling based on tumor response. A, Quantification of PD-L1, ILF3, STAT3, and pSTAT3 showed increased ILF3 and STAT3 expression in nonresponding tumors. NR, nonresponders; R, responders. B, Representative IHC images of ILF3 and STAT3 staining by response type. C, All components of the PD-L1 interactome were increased in early compared with late DR cases. Five representative images per case were quantified; lines represent mean, and errors bars represent standard error of the mean. D, Top, differences in STAT3 expression by treatment response were specific to tissue samples collected just prior to ICI initiation. E, Multicolor PLA representative images (left) and quantification (right) revealed significantly higher PD-L1 intracellular signaling activity in nonresponding vs. responding cases as demonstrated by increased interaction between PD-L1–ILF3 and PD-L1–STAT3. Five cases per group with quantification from five representative images. Scale bars, 10 μm (left); magnified PLA is 4× original size.
- Open Access
- 8 October 2026
- Published by American Association for Cancer Research (AACR)
Supplemental Table 2 from Dissociated Response to Immune Checkpoint Inhibition in Head and Neck Squamous Cell Carcinoma Is Common and Exhibits Distinct Clinical Outcomes
Supplemental Table 2. Response by disease site in patients with DR.
- 8 October 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplemental Table 2 from Dissociated Response to Immune Checkpoint Inhibition in Head and Neck Squamous Cell Carcinoma Is Common and Exhibits Distinct Clinical Outcomes
Supplemental Table 2. Response by disease site in patients with DR.
- Open Access
- 8 October 2026
- Published by American Association for Cancer Research (AACR)
Supplemental Figure 1 from Dissociated Response to Immune Checkpoint Inhibition in Head and Neck Squamous Cell Carcinoma Is Common and Exhibits Distinct Clinical Outcomes
Supplemental Figure 1. Timing of treatment responses in patients with DR. Timing of best response (red), dissociated response (DR, orange), and overall progression of disease (PD, blue) in months from the time of first immunotherapy cycle in patients with dissociated response. Those patients where best response and dissociated response occurred at the same time were classified as early DR (grouped at the top), and those patients where dissociated response occurred at least 1 month after best response were classified as late DR (grouped at the bottom).
- 8 October 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplemental Figure 1 from Dissociated Response to Immune Checkpoint Inhibition in Head and Neck Squamous Cell Carcinoma Is Common and Exhibits Distinct Clinical Outcomes
Supplemental Figure 1. Timing of treatment responses in patients with DR. Timing of best response (red), dissociated response (DR, orange), and overall progression of disease (PD, blue) in months from the time of first immunotherapy cycle in patients with dissociated response. Those patients where best response and dissociated response occurred at the same time were classified as early DR (grouped at the top), and those patients where dissociated response occurred at least 1 month after best response were classified as late DR (grouped at the bottom).
- Open Access
- 8 October 2026
- Published by American Association for Cancer Research (AACR)
Supplemental Figure 2 from Dissociated Response to Immune Checkpoint Inhibition in Head and Neck Squamous Cell Carcinoma Is Common and Exhibits Distinct Clinical Outcomes
Supplemental Figure 2. Survival outcomes based on lung-specific treatment response. Kaplan-Meier estimates of progression free survival (PFS, left) and overall survival (OS, right) in months based on organ-specific treatment response of lung metastases. Tick marks indicate censored data. Abbreviations: complete response (CR), not reached (NR), partial response (PR), stable disease (SD), and progressive disease (PD).
- 8 October 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplemental Figure 2 from Dissociated Response to Immune Checkpoint Inhibition in Head and Neck Squamous Cell Carcinoma Is Common and Exhibits Distinct Clinical Outcomes
Supplemental Figure 2. Survival outcomes based on lung-specific treatment response. Kaplan-Meier estimates of progression free survival (PFS, left) and overall survival (OS, right) in months based on organ-specific treatment response of lung metastases. Tick marks indicate censored data. Abbreviations: complete response (CR), not reached (NR), partial response (PR), stable disease (SD), and progressive disease (PD).