Habitat escalated adaptive therapy (HEAT): a phase 2 trial utilizing radiomic habitat-directed and genomic-adjusted radiation dose (GARD) optimization for high-grade soft tissue sarcoma.

BACKGROUND: Soft tissue sarcomas (STS), have significant inter- and intra-tumoral heterogeneity, with poor response to standard neoadjuvant radiotherapy (RT). Achieving a favorable pathologic response (FPR ≥ 95%) from RT is associated with improved patient outcome. Genomic adjusted radiation dose (GARD), a radiation-specific metric that quantifies the expected RT treatment effect as a function of tumor dose and genomics, proposed that STS is significantly underdosed. STS have significant radiomic heterogeneity, where radiomic habitats can delineate regions of intra-tumoral hypoxia and radioresistance. We designed a novel clinical trial, Habitat Escalated Adaptive Therapy (HEAT), utilizing radiomic habitats to identify areas of radioresistance within the tumor and targeting them with GARD-optimized doses, to improve FPR in high-grade STS. METHODS: Phase 2 non-randomized single-arm clinical trial includes non-metastatic, resectable high-grade STS patients. Pre-treatment multiparametric MRIs (mpMRI) delineate three distinct intra-tumoral habitats based on apparent diffusion coefficient (ADC) and dynamic contrast enhanced (DCE) sequences. GARD estimates that simultaneous integrated boost (SIB) doses of 70 and 60 Gy in 25 fractions to the highest and intermediate radioresistant habitats, while the remaining volume receives standard 50 Gy, would lead to a > 3 fold FPR increase to 24%. Pre-treatment CT guided biopsies of each habitat along with clip placement will be performed for pathologic evaluation, future genomic studies, and response assessment. An mpMRI taken between weeks two and three of treatment will be used for biological plan adaptation to account for tumor response, in addition to an mpMRI after the completion of radiotherapy in addition to pathologic response, toxicity, radiomic response, disease control, and survival will be evaluated as secondary endpoints. Furthermore, liquid biopsy will be performed with mpMRI for future ancillary studies. DISCUSSION: This is the first clinical trial to test a novel genomic-based RT dose optimization (GARD) and to utilize radiomic habitats to identify and target radioresistance regions, as a strategy to improve the outcome of RT-treated STS patients. Its success could usher in a new phase in radiation oncology, integrating genomic and radiomic insights into clinical practice and trial designs, and may reveal new radiomic and genomic biomarkers, refining personalized treatment strategies for STS. TRIAL REGISTRATION: NCT05301283. TRIAL STATUS: The trial started recruitment on March 17, 2022.

Pan-cancer prediction of radiotherapy benefit using genomic-adjusted radiation dose (GARD): a cohort-based pooled analysis.

BACKGROUND: Despite advances in cancer genomics, radiotherapy is still prescribed on the basis of an empirical one-size-fits-all paradigm. Previously, we proposed a novel algorithm using the genomic-adjusted radiation dose (GARD) model to personalise prescription of radiation dose on the basis of the biological effect of a given physical dose of radiation, calculated using individual tumour genomics. We hypothesise that GARD will reveal interpatient heterogeneity associated with opportunities to improve outcomes compared with physical dose of radiotherapy alone. We aimed to test this hypothesis and investigate the GARD-based radiotherapy dosing paradigm. METHODS: We did a pooled, pan-cancer analysis of 11 previously published clinical cohorts of unique patients with seven different types of cancer, which are all available cohorts with the data required to calculate GARD, together with clinical outcome. The included cancers were breast cancer, head and neck cancer, non-small-cell lung cancer, pancreatic cancer, endometrial cancer, melanoma, and glioma. Our dataset comprised 1615 unique patients, of whom 1298 (982 with radiotherapy, 316 without radiotherapy) were assessed for time to first recurrence and 677 patients (424 with radiotherapy and 253 without radiotherapy) were assessed for overall survival. We analysed two clinical outcomes of interest: time to first recurrence and overall survival. We used Cox regression, stratified by cohort, to test the association between GARD and outcome with separate models using dose of radiation and sham-GARD (ie, patients treated without radiotherapy, but modelled as having a standard-of-care dose of radiotherapy) for comparison. We did interaction tests between GARD and treatment (with or without radiotherapy) using the Wald statistic. FINDINGS: Pooled analysis of all available data showed that GARD as a continuous variable is associated with time to first recurrence (hazard ratio [HR] 0·98 [95% CI 0·97-0·99]; p=0·0017) and overall survival (0·97 [0·95-0·99]; p=0·0007). The interaction test showed the effect of GARD on overall survival depends on whether or not that patient received radiotherapy (Wald statistic p=0·011). The interaction test for GARD and radiotherapy was not significant for time to first recurrence (Wald statistic p=0·22). The HR for physical dose of radiation was 0·99 (95% CI 0·97-1·01; p=0·53) for time to first recurrence and 1·00 (0·96-1·04; p=0·95) for overall survival. The HR for sham-GARD was 1·00 (0·97-1·03; p=1·00) for time to first recurrence and 1·00 (0·98-1·02; p=0·87) for overall survival. INTERPRETATION: The biological effect of radiotherapy, as quantified by GARD, is significantly associated with time to first recurrence and overall survival for patients with cancer treated with radiation. It is predictive of radiotherapy benefit, and physical dose of radiation is not. We propose integration of genomics into radiation dosing decisions, using a GARD-based framework, as the new paradigm for personalising radiotherapy prescription dose. FUNDING: None. VIDEO ABSTRACT.

Effectiveness and Safety of Prostatic Artery Embolization for the Treatment of Lower Urinary Tract Symptoms from Benign Prostatic Hyperplasia in Men with Concurrent Localized Prostate Cancer.

PURPOSE: To assess the effectiveness and safety of prostatic artery embolization (PAE) on lower urinary tract symptoms (LUTS) in the setting of localized prostate cancer (PCa). MATERIALS AND METHODS: This was a retrospective, single-center, institutional review board-approved study from December 2016 to June 2020 of 21 patients (median age, 72; range, 63-83 years) with moderate LUTS and localized PCa. Clinical effectiveness was evaluated at 6 and 12 weeks using International Prostate Symptom Score (IPSS) and quality of life (QoL) improvement. Seventeen patients were scheduled to receive definitive radiotherapy (RT) after PAE; 13 patients completed RT. Short-term imaging signs of oncologic progression were evaluated at 6 and 12 weeks defined by at least one of the following on magnetic resonance imaging: increased Prostate Imaging-Reporting and Data System score of index lesion(s) to at least 4, new extracapsular extension, seminal vesicle involvement, or pelvic lymphadenopathy. Nonparametric Wilcoxon signed-rank test was used for analysis. RESULTS: IPSS improved by a median of 12 (n = 19, P < .0001) and 14 (n = 14, P < .0001) at 6 and 12 weeks, respectively. QoL improved by a median of 2 (n = 19, P < .0001) and 3 (n = 3, P < .0001) at 6 and 12 weeks. Prostate volume decreased by a median of 24% (n = 19, P < .0001) and 36% (n = 12, P = .015) at 6 and 12 weeks. No patients demonstrated disease progression at 6 (n = 16) or 12 (n = 8) weeks by imaging. No patients experienced increased prostate-specific antigen after RT, grade ≥3 adverse events, or greater genitourinary toxicity. CONCLUSIONS: PAE is effective and safe for the treatment of men with LUTS from benign prostatic hyperplasia in the setting of concomitant, localized, non-obstructive PCa.

Modeling precision genomic-based radiation dose response in rectal cancer.

Aim: Genomic-based risk stratification to personalize radiation dose in rectal cancer. Patients & methods: We modeled genomic-based radiation dose response using the previously validated radiosensitivity index (RSI) and the clinically actionable genomic-adjusted radiation dose. Results: RSI of rectal cancer ranged from 0.19 to 0.81 in a bimodal distribution. A pathologic complete response rate of 21% was achieved in tumors with an RSI <0.31 at a minimal genomic-adjusted radiation dose of 29.76 when modeling RxRSI to the commonly prescribed physical dose of 50 Gy. RxRSI-based dose escalation to 55 Gy in tumors with an RSI of 0.31-0.34 could increase pathologic complete response by 10%. Conclusion: This study provides a theoretical platform for development of an RxRSI-based prospective trial in rectal cancer.

Management of Sentinel Lymph Node Metastasis in Merkel Cell Carcinoma: Completion Lymphadenectomy, Radiation, or Both?

BACKGROUND: Approximately 30% of patients with clinically localized Merkel cell carcinoma (MCC) show nodal involvement on sentinel lymph node biopsy (SLNB). Optimal management of SLNB-positive disease has not been defined. This study compared outcomes after completion lymphadenectomy (CLND), radiation, and combined CLND plus radiation after a positive SLNB. METHODS: All patients treated at a single institution for SLNB-positive MCC (1998-2015) were retrospectively evaluated, with examination of patient demographics, clinicopathologic characteristics, outcomes, and regional toxicity. RESULTS: The study identified 71 evaluable patients with SLNB-positive disease. The median age of these patients was 76 years, and 76.1% were men. Of the 71 patients, 11 (15.5%) underwent CLND, 40 (56.3%) received radiation, and 20 (28.2%) underwent CLND plus postoperative radiation. Lymphovascular invasion was significantly more common in the radiation-alone cohort (p = 0.04). For the three cohorts, the median percentages of nodal involvement were respectively 2, 10, and 30% (p = 0.06). After a median follow-up period of 22.3 months, four patients had recurrence in their regional nodal basin (3 radiation-alone patients and 1 CLND + radiation patient). The three cohorts did not differ significantly in the development of distant metastases (p = 0.68) or overall survival (p = 0.72). Six patients experienced surgical-site infections (2 CLND and 4 CLND + radiation patients), and three patients experienced symptomatic lymphedema (1 CLND patient and 2 CLND + radiation patients). CONCLUSIONS: Regional failure was infrequent (≤ 10%) regardless of treatment, and morbidity appeared to be low with all approaches. Given that multiple treatment approaches can be successful in treating micrometastatic MCC, future efforts should be directed at refining criteria for allocating patients to a specific method, or possibly no further nodal basin treatment, in an effort to maximize regional control at the lowest cost and morbidity.

Intrinsic radiosensitivity, genomic-based radiation dose and patterns of failure of penile cancer in response to adjuvant radiation therapy.

PURPOSE: Optimal postoperative radiation therapy (PORT) dose is unclear in penile squamous cell carcinoma (PeSCC). Herein, we characterized the radiosensitivity index (RSI) and genomic-adjusted radiation dose (GARD) profiles in a cohort of patients with PeSCC, and assessed the application of GARD to personalize PORT. METHODS: A total of 25 PeSCC samples were identified for transcriptomic profiling. The RSI score and GARD were derived for each sample. A cohort of 34 patients was reviewed for clinical correlation. RESULTS: The median RSI for PeSCC was 0.482 (range 0.215-0.682). The majority (n = 21; 84%) of cases were classified as radioresistant. PeSCC GARD ranged from 9.56 to 38.39 (median 18.25), suggesting variable therapeutic effects from PORT. We further determined the optimal GARD-based RT doses to improve locoregional control. We found that therapeutic benefit was only achieved in 52% of PeSCC lesions with PORT of 50 Gy, in contrast to 84% benefit from GARD-modeled PORT of 66 Gy. In the clinical cohort, the majority of patients presented with pathological N2 or N3 disease (n = 31; 91%) and was treated with adjuvant concurrent platinum-based chemoradiotherapy (CRT, n = 30; 88%). Fourteen of the 34 patients (41%) had locoregional recurrence (LRR), of which half had LRR within six months of completion of PORT. CONCLUSIONS: The majority of PeSCC are intrinsically radioresistant with a low GARD-based therapeutic effect from PORT dose of 50 Gy, consistent with the observed high rate of LRR in the clinical cohort. A GARD-based strategy will allow personalizing PORT dose prescription to individual tumor biology and improve outcomes.

The future of personalised radiotherapy for head and neck cancer.

Radiotherapy has long been the mainstay of treatment for patients with head and neck cancer and has traditionally involved a stage-dependent strategy whereby all patients with the same TNM stage receive the same therapy. We believe there is a substantial opportunity to improve radiotherapy delivery beyond just technological and anatomical precision. In this Series paper, we explore several new ideas that could improve understanding of the phenotypic and genotypic differences that exist between patients and their tumours. We discuss how exploiting these differences and taking advantage of precision medicine tools-such as genomics, radiomics, and mathematical modelling-could open new doors to personalised radiotherapy adaptation and treatment. We propose a new treatment shift that moves away from an era of empirical dosing and fractionation to an era focused on the development of evidence to guide personalisation and biological adaptation of radiotherapy. We believe these approaches offer the potential to improve outcomes and reduce toxicity.

A genome-based model for adjusting radiotherapy dose (GARD): a retrospective, cohort-based study.

BACKGROUND: Despite its common use in cancer treatment, radiotherapy has not yet entered the era of precision medicine, and there have been no approaches to adjust dose based on biological differences between or within tumours. We aimed to assess whether a patient-specific molecular signature of radiation sensitivity could be used to identify the optimum radiotherapy dose. METHODS: We used the gene-expression-based radiation-sensitivity index and the linear quadratic model to derive the genomic-adjusted radiation dose (GARD). A high GARD value predicts for high therapeutic effect for radiotherapy; which we postulate would relate to clinical outcome. Using data from the prospective, observational Total Cancer Care (TCC) protocol, we calculated GARD for primary tumours from 20 disease sites treated using standard radiotherapy doses for each disease type. We also used multivariable Cox modelling to assess whether GARD was independently associated with clinical outcome in five clinical cohorts: Erasmus Breast Cancer Cohort (n=263); Karolinska Breast Cancer Cohort (n=77); Moffitt Lung Cancer Cohort (n=60); Moffitt Pancreas Cancer Cohort (n=40); and The Cancer Genome Atlas Glioblastoma Patient Cohort (n=98). FINDINGS: We calculated GARD for 8271 tissue samples from the TCC cohort. There was a wide range of GARD values (range 1·66-172·4) across the TCC cohort despite assignment of uniform radiotherapy doses within disease types. Median GARD values were lowest for gliomas and sarcomas and highest for cervical cancer and oropharyngeal head and neck cancer. There was a wide range of GARD values within tumour type groups. GARD independently predicted clinical outcome in breast cancer, lung cancer, glioblastoma, and pancreatic cancer. In the Erasmus Breast Cancer Cohort, 5-year distant-metastasis-free survival was longer in patients with high GARD values than in those with low GARD values (hazard ratio 2·11, 95% 1·13-3·94, p=0·018). INTERPRETATION: A GARD-based clinical model could allow the individualisation of radiotherapy dose to tumour radiosensitivity and could provide a framework to design genomically-guided clinical trials in radiation oncology. FUNDING: None.

Regional Radiation Therapy Impacts Outcome for Node-Positive Cutaneous Melanoma.

Background: Regional radiation therapy (RT) has been shown to reduce the risk of regional recurrence with node-positive cutaneous melanoma. However, risk factors for regional recurrence, especially in the era of sentinel lymph node biopsy (SLNB), are less clear. Our goals were to identify risk factors associated with regional recurrence and to determine whether a radiosensitivity index (RSI) gene expression signature (GES) could identify patients who experience a survival benefit with regional RT. Methods: A single-institution, Institutional Review Board-approved study was performed including 410 patients treated with either SLNB with or without completion lymph node dissection (LND; n=270) or therapeutic LND (n=91). Postoperative regional RT was delivered to the involved nodal basin in 83 cases (20.2%), to a median dose of 54 Gy (range, 30-60 Gy) in 27 fractions (range, 5-30). Primary outcomes were regional control and overall survival by RSI GES status. Results: Median follow-up was 69 months (range, 13-180). Postoperative regional RT was associated with a reduced risk of regional recurrence among all patients on univariate (5-year estimate: 95.0% vs 83.3%; P=.036) and multivariate analysis (hazard ratio[HR], 0.15; 95% CI, 0.05-0.43; P<.001). Among higher-risk subgroups, regional RT was associated with a lower risk of regional recurrence among patients with clinically detected lymph nodes (n=175; 5-year regional control: 94.1% vs 69.5%; P=.003) and extracapsular extension (ECE) present (n=138; 5-year regional control: 96.7% vs 62.2%; P<.001). Among a subset of radiated patients with gene expression data available, a low RSI GES (radiosensitive) tumor status was associated with improved survival compared with a high RSI GES (5-year: 75% vs 0%; HR, 10.68; 95% CI, 1.24-92.14). Conclusions: Regional RT was associated with a reduced risk of regional recurrence among patients with ECE and clinically detected nodal disease. Gene expression data show promise for better predicting radiocurable patients in the future. In the era of increasingly effective systemic therapies, the value of improved regional control potentially takes on greater significance.

Radiation Therapy is Associated with Improved Outcomes in Merkel Cell Carcinoma.

BACKGROUND: Following wide excision of Merkel cell carcinoma (MCC), postoperative radiation therapy (RT) is typically recommended. Controversy remains as to whether RT can be avoided in selected cases, such as those with negative margins. Additionally, there is evidence that RT can influence survival. METHODS: We included 171 patients treated for non-metastatic MCC from 1994 through 2012 at a single institution. Patients without pathologic nodal evaluation (clinical N0 disease) were excluded to reflect modern treatment practice. The endpoints included local control (LC), locoregional control (LRC), disease-free survival (DFS), overall survival (OS), and disease-specific survival (DSS). RESULTS: Median follow-up was 33 months. Treatment with RT was associated with improved 3-year LC (91.2 vs. 76.9 %, respectively; p = 0.01), LRC (79.5 vs. 59.1 %; p = 0.004), DFS (57.0 vs. 30.2 %; p < 0.001), and OS (73 vs. 66 %; p = 0.02), and was associated with improved 3-year DSS among node-positive patients (76.2 vs. 48.1 %; p = 0.035), but not node-negative patients (90.1 vs. 80.8 %; p = 0.79). On multivariate analysis, RT was associated with improved LC [hazard ratio (HR) 0.18, 95 % confidence interval (CI) 0.07-0.46; p < 0.001], LRC (HR 0.28, 95 % CI 0.14-0.56; p < 0.001), DFS (HR 0.42, 95 % CI 0.26-0.70; p = 0.001), OS (HR 0.53, 95 % CI 0.31-0.93; p = 0.03), and DSS (HR 0.42, 95 % CI 0.26-0.70; p = 0.001). Patients with negative margins had significant improvements in 3-year LC (90.1 vs. 75.4 %; p < 0.001) with RT. Deaths not attributable to MCC were relatively evenly distributed between the RT and no RT groups (28.5 and 29.3 % of patients, respectively). CONCLUSIONS: RT for MCC was associated with improved LRC and survival. RT appeared to be beneficial regardless of margin status.

Melanoma, Version 2.2016, NCCN Clinical Practice Guidelines in Oncology.

This selection from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines) for Melanoma focuses on adjuvant therapy and treatment of in-transit disease, because substantial changes were made to the recommendations for the 2016 update. Depending on the stage of the disease, options for adjuvant therapy now include biochemotherapy and high-dose ipilimumab. Treatment options for in-transit disease now include intralesional injection with talimogene laherparepvec (T-VEC), a new immunotherapy. These additions prompted re-assessment of the data supporting older recommended treatment options for adjuvant therapy and in-transit disease, resulting in extensive revisions to the supporting discussion sections.

NCCN Guidelines Insights: Melanoma, Version 3.2016.

The NCCN Guidelines for Melanoma have been significantly revised over the past few years in response to emerging data on a number of novel agents and treatment regimens. These NCCN Guidelines Insights summarize the data and rationale supporting extensive changes to the recommendations for systemic therapy in patients with metastatic or unresectable melanoma.

Stereotactic Body Radiotherapy in the Management of Oligometastatic Disease.

BACKGROUND: The treatment of oligometastatic disease has become common as imaging techniques have advanced and the management of systemic disease has improved. Use of highly targeted, hypofractionated regimens of stereotactic body radiotherapy (SBRT) is now a primary management option for patients with oligometastatic disease. METHODS: The properties of SBRT are summarized and the results of retrospective and prospective studies of SBRT use in the management of oligometastases are reviewed. Future directions of SBRT, including optimizing dose and fractionation schedules, are also discussed. RESULTS: SBRT can deliver highly conformal, dosed radiation treatments for ablative tumors in a few treatment sessions. Phase 1/2 trials and retrospective institutional results support use of SBRT as a treatment option for oligometastatic disease metastasized to the lung, liver, and spine, and SBRT offers adequate toxicity profiles with good rates of local control. Future directions will involve optimizing dose and fractionation schedules for select histologies to improve rates of local control while limiting toxicity to normal structures. CONCLUSIONS: SBRT offers an excellent management option for patients with oligometastases. However, additional research is still needed to optimize dose and fractionation schedules.

Using the genomic adjusted radiation dose (GARD) to personalize the radiation dose in nasopharyngeal cancer.

BACKGROUND: Locally advanced nasopharyngeal cancer (NPC) patients undergoing radiotherapy are at risk of treatment failure, particularly locoregional recurrence. To optimize the individual radiation dose, we hypothesize that the genomic adjusted radiation dose (GARD) can be used to correlate with locoregional control. METHODS: A total of 92 patients with American Joint Committee on Cancer / International Union Against Cancer stage III to stage IVB recruited in a randomized phase III trial were assessed (NPC-0501) (NCT00379262). Patients were treated with concurrent chemo-radiotherapy plus (neo) adjuvant chemotherapy. The primary endpoint is locoregional failure free rate (LRFFR). RESULTS: Despite the homogenous physical radiation dose prescribed (Median: 70 Gy, range 66-76 Gy), there was a wide range of GARD values (median: 50.7, range 31.1-67.8) in this cohort. In multivariable analysis, a GARD threshold (GARDT) of 45 was independently associated with LRFFR (p = 0.008). By evaluating the physical dose required to achieve the GARDT (RxRSI), three distinct clinical subgroups were identified: (1) radiosensitive tumors that RxRSI at dose < 66 Gy (N = 59, 64.1 %) (b) moderately radiosensitive tumors that RxRSI dose within the current standard of care range (66-74 Gy) (N = 20, 21.7 %), (c) radioresistant tumors that need a significant dose escalation above the current standard of care (>74 Gy) (N = 13, 14.1 %). CONCLUSION: GARD is independently associated with locoregional control in radiotherapy-treated NPC patients from a Phase 3 clinical trial. GARD may be a potential framework to personalize radiotherapy dose for NPC patients.

Changes in Daily Apparent Diffusion Coefficient on Fully Quantitative Magnetic Resonance Imaging Correlate With Established Genomic Pathways of Radiation Sensitivity and Reveal Novel Biologic Associations.

PURPOSE: Changes in quantitative magnetic resonance imaging (qMRI) are frequently observed during chemotherapy or radiation therapy (RT). It is hypothesized that qMRI features are reflective of underlying tissue responses. It's unknown what underlying genomic characteristics underly qMRI changes. We hypothesized that qMRI changes may correlate with DNA damage response (DDR) capacity within human tumors. Therefore, we designed the current study to correlate qMRI changes from daily RT treatment with underlying tumor transcriptomic profiles. METHODS AND MATERIALS: Study participants were prospectively enrolled (National Clinical Trial 03500081). RNA expression levels for 757 genes from pretreatment biopsies were obtained using a custom panel that included signatures of radiation sensitivity and DDR. Daily qMRI data were obtained from a 1.5 Tesla MR linear accelerator. Using these images, d-slow, d-star, perfusion, and apparent diffusion coefficient-mean values in tumors were plotted per-fraction, over time, and associated with genomic pathways. RESULTS: A total of 1022 qMRIs were obtained from 39 patients and both genomic data and qMRI data from 27 total patients. For 20 of those patients, we also generated normal tissue transcriptomic data. Radio sensitivity index values most closely associated with tissue of origin. Multiple genomic pathways including DNA repair, peroxisome, late estrogen receptor responses, KRAS signaling, and UV response were significantly associated with qMRI feature changes (P < .001). CONCLUSIONS: Genomic pathway associations across metabolic, RT sensitivity, and DDR pathways indicate common tumor biology that may correlate with qMRI changes during a course of treatment. Such data provide hypothesis-generating novel mechanistic insight into the biologic meaning of qMRI changes during treatment and enable optimal selection of imaging biomarkers for biologically MR-guided RT.

Intricacies of Human-AI Interaction in Dynamic Decision-Making for Precision Oncology: A Case Study in Response-Adaptive Radiotherapy.

Background: Adaptive treatment strategies that can dynamically react to individual cancer progression can provide effective personalized care. Longitudinal multi-omics information, paired with an artificially intelligent clinical decision support system (AI-CDSS) can assist clinicians in determining optimal therapeutic options and treatment adaptations. However, AI-CDSS is not perfectly accurate, as such, clinicians' over/under reliance on AI may lead to unintended consequences, ultimately failing to develop optimal strategies. To investigate such collaborative decision-making process, we conducted a Human-AI interaction case study on response-adaptive radiotherapy (RT). Methods: We designed and conducted a two-phase study for two disease sites and two treatment modalities-adaptive RT for non-small cell lung cancer (NSCLC) and adaptive stereotactic body RT for hepatocellular carcinoma (HCC)-in which clinicians were asked to consider mid-treatment modification of the dose per fraction for a number of retrospective cancer patients without AI-support (Unassisted Phase) and with AI-assistance (AI-assisted Phase). The AI-CDSS graphically presented trade-offs in tumor control and the likelihood of toxicity to organs at risk, provided an optimal recommendation, and associated model uncertainties. In addition, we asked for clinicians' decision confidence level and trust level in individual AI recommendations and encouraged them to provide written remarks. We enrolled 13 evaluators (radiation oncology physicians and residents) from two medical institutions located in two different states, out of which, 4 evaluators volunteered in both NSCLC and HCC studies, resulting in a total of 17 completed evaluations (9 NSCLC, and 8 HCC). To limit the evaluation time to under an hour, we selected 8 treated patients for NSCLC and 9 for HCC, resulting in a total of 144 sets of evaluations (72 from NSCLC and 72 from HCC). Evaluation for each patient consisted of 8 required inputs and 2 optional remarks, resulting in up to a total of 1440 data points. Results: AI-assistance did not homogeneously influence all experts and clinical decisions. From NSCLC cohort, 41 (57%) decisions and from HCC cohort, 34 (47%) decisions were adjusted after AI assistance. Two evaluations (12%) from the NSCLC cohort had zero decision adjustments, while the remaining 15 (88%) evaluations resulted in at least two decision adjustments. Decision adjustment level positively correlated with dissimilarity in decision-making with AI [NSCLC: ρ = 0.53 ( p < 0.001); HCC: ρ = 0.60 ( p < 0.001)] indicating that evaluators adjusted their decision closer towards AI recommendation. Agreement with AI-recommendation positively correlated with AI Trust Level [NSCLC: ρ = 0.59 ( p < 0.001); HCC: ρ = 0.7 ( p < 0.001)] indicating that evaluators followed AI's recommendation if they agreed with that recommendation. The correlation between decision confidence changes and decision adjustment level showed an opposite trend [NSCLC: ρ = -0.24 ( p = 0.045), HCC: ρ = 0.28 ( p = 0.017)] reflecting the difference in behavior due to underlying differences in disease type and treatment modality. Decision confidence positively correlated with the closeness of decisions to the standard of care (NSCLC: 2 Gy/fx; HCC: 10 Gy/fx) indicating that evaluators were generally more confident in prescribing dose fractionations more similar to those used in standard clinical practice. Inter-evaluator agreement increased with AI-assistance indicating that AI-assistance can decrease inter-physician variability. The majority of decisions were adjusted to achieve higher tumor control in NSCLC and lower normal tissue complications in HCC. Analysis of evaluators' remarks indicated concerns for organs at risk and RT outcome estimates as important decision-making factors. Conclusions: Human-AI interaction depends on the complex interrelationship between expert's prior knowledge and preferences, patient's state, disease site, treatment modality, model transparency, and AI's learned behavior and biases. The collaborative decision-making process can be summarized as follows: (i) some clinicians may not believe in an AI system, completely disregarding its recommendation, (ii) some clinicians may believe in the AI system but will critically analyze its recommendations on a case-by-case basis; (iii) when a clinician finds that the AI recommendation indicates the possibility for better outcomes they will adjust their decisions accordingly; and (iv) When a clinician finds that the AI recommendation indicate a worse possible outcome they will disregard it and seek their own alternative approach.

Exploiting convergent phenotypes to derive a pan-cancer cisplatin response gene expression signature.

Precision medicine offers remarkable potential for the treatment of cancer, but is largely focused on tumors that harbor actionable mutations. Gene expression signatures can expand the scope of precision medicine by predicting response to traditional (cytotoxic) chemotherapy agents without relying on changes in mutational status. We present a new signature extraction method, inspired by the principle of convergent phenotypes, which states that tumors with disparate genetic backgrounds may evolve similar phenotypes independently. This evolutionary-informed method can be utilized to produce consensus signatures predictive of response to over 200 chemotherapeutic drugs found in the Genomics of Drug Sensitivity in Cancer (GDSC) Database. Here, we demonstrate its use by extracting the Cisplatin Response Signature (CisSig). We show that this signature can predict cisplatin response within carcinoma-based cell lines from the GDSC database, and expression of the signatures aligns with clinical trends seen in independent datasets of tumor samples from The Cancer Genome Atlas (TCGA) and Total Cancer Care (TCC) database. Finally, we demonstrate preliminary validation of CisSig for use in muscle-invasive bladder cancer, predicting overall survival in a small cohort of patients who undergo cisplatin-containing chemotherapy. This methodology can be used to produce robust signatures that, with further clinical validation, may be used for the prediction of traditional chemotherapeutic response, dramatically increasing the reach of personalized medicine in cancer.