Harnessing the Potential of Real-World Evidence in the Treatment of Colorectal Cancer: Where Do We Stand?

OPINION STATEMENT: Treatment guidelines for colorectal cancer (CRC) are primarily based on the results of randomized clinical trials (RCTs), the gold standard methodology to evaluate safety and efficacy of oncological treatments. However, generalizability of trial results is often limited due to stringent eligibility criteria, underrepresentation of specific populations, and more heterogeneity in clinical practice. This may result in an efficacy-effectiveness gap and uncertainty regarding meaningful benefit versus treatment harm. Meanwhile, conduct of traditional RCTs has become increasingly challenging due to identification of a growing number of (small) molecular subtypes. These challenges-combined with the digitalization of health records-have led to growing interest in use of real-world data (RWD) to complement evidence from RCTs. RWD is used to evaluate epidemiological trends, quality of care, treatment effectiveness, long-term (rare) safety, and quality of life (QoL) measures. In addition, RWD is increasingly considered in decision-making by clinicians, regulators, and payers. In this narrative review, we elaborate on these applications in CRC, and provide illustrative examples. As long as the quality of RWD is safeguarded, ongoing developments, such as common data models, federated learning, and predictive modelling, will further unfold its potential. First, whenever possible, we recommend conducting pragmatic trials, such as registry-based RCTs, to optimize generalizability and answer clinical questions that are not addressed in registrational trials. Second, we argue that marketing approval should be conditional for patients who would have been ineligible for the registrational trial, awaiting planned (non) randomized evaluation of outcomes in the real world. Third, high-quality effectiveness results should be incorporated in treatment guidelines to aid in patient counseling. We believe that a coordinated effort from all stakeholders is essential to improve the quality of RWD, create a learning healthcare system with optimal use of trials and real-world evidence (RWE), and ultimately ensure personalized care for every CRC patient.

Survival of Patients with Deficient Mismatch Repair Versus Proficient Mismatch Repair Metastatic Colorectal Cancer Receiving Curative-Intent Local Treatment of Metastases in a Nationwide Cohort.

BACKGROUND: It is unclear whether curative-intent local therapy of metastases is of similar benefit for the biological distinct subgroup of patients with deficient mismatch repair (dMMR) metastatic colorectal cancer (mCRC) compared with proficient mismatch repair (pMMR) mCRC. PATIENTS AND METHODS: In this nationwide study, recurrence-free (RFS) and overall survival (OS) were analyzed in patients with dMMR versus pMMR mCRC who underwent curative-intent local treatment of metastases between 2015 and 2018. Subgroup analyses were performed for resection of colorectal liver metastases (CRLM) and cytoreductive surgery ± hyperthermic intraperitoneal chemotherapy (CRS ± HIPEC). Multivariable regression was conducted. RESULTS: Median RFS was 11.1 months [95% confidence interval (CI) 8.5-41.1 months] for patients with dMMR tumors compared with 8.9 months (95% CI 8.1-9.8 months) for pMMR tumors. Two-year RFS was higher in patients with dMMR versus pMMR (43% vs. 21%). Results were similar within subgroups of local treatment (CRLM and CRS ± HIPEC). Characteristics differed significantly between patients with dMMR and pMMR mCRC; however, multivariable analysis continued to demonstrate dMMR as independent factor for improved RFS [hazard ratio (HR): 0.57, 95% CI 0.38-0.87]. Median OS was 33.3 months for dMMR mCRC compared with 43.5 months for pMMR mCRC, mainly due to poor survival of patients with dMMR in cases of recurrence in the preimmunotherapy era. CONCLUSION: Patients with dMMR eligible for curative-intent local treatment of metastases showed a comparable to more favorable RFS compared with patients with pMMR, with a clinically relevant proportion of patients remaining free of recurrence. This supports local treatment as a valuable treatment option in patients with dMMR mCRC and can aid in shared decision-making regarding upfront local therapy versus immunotherapy.

Prognostic value of Lynch syndrome, BRAFV600E , and RAS mutational status in dMMR/MSI-H metastatic colorectal cancer in a pooled analysis of Dutch and French cohorts.

BACKGROUND: Current knowledge on prognostic biomarkers (especially BRAFV600E /RAS mutations) in metastatic colorectal cancer (mCRC) is mainly based on mCRC patients with proficient mismatch repair (pMMR) tumors. It is uncertain whether these biomarkers have the same prognostic value in mCRC patients with deficient mismatch repair (dMMR) tumors. METHODS: This observational cohort study combined a population-based Dutch cohort (2014-2019) and a large French multicenter cohort (2007-2017). All mCRC patients with a histologically proven dMMR tumor were included. RESULTS: In our real-world data cohort of 707 dMMR mCRC patients, 438 patients were treated with first-line palliative systemic chemotherapy. Mean age of first-line treated patients was 61.9 years, 49% were male, and 40% had Lynch syndrome. BRAFV600E mutation was present in 47% of tumors and 30% harbored a RAS mutation. Multivariable regression analysis on OS showed significant hazard rates (HR) for known prognostic factors as age and performance status, however showed no significance for Lynch syndrome (HR: 1.07, 95% CI: 0.66-1.72), BRAFV600E mutational status (HR: 1.02, 95% CI: 0.67-1.54), and RAS mutational status (HR: 1.01, 95% CI: 0.64-1.59), with similar results for PFS. CONCLUSION: BRAFV600E and RAS mutational status are not associated with prognosis in dMMR mCRC patients, in contrast to pMMR mCRC patients. Lynch syndrome is also not an independent prognostic factor for survival. These findings underline that prognostic factors of patients with dMMR mCRC are different of those with pMMR, which could be taken into consideration when prognosis is used for clinical decision-making in dMMR mCRC patients and underline the complex heterogeneity of mCRC.

Use of preventive medication and supplements in general practice in patients in their last year of life: a Retrospective cohort study.

BACKGROUND: Several preventive medications and supplements become inappropriate in the last phase of life due to increased risk of adverse events caused by changed pharmacokinetics, drug-drug interactions, and changed care goals. Information on these preventive medication and supplements use in patients with a life-limiting illness in the home-care setting is limited. The primary aim of this study was to assess the use of four different groups of preventive drugs and supplements, which are inappropriate in adult patients with a life-limiting illness, living at home in the last year of life. The secondary aims were to assess reasons for discontinuing these drugs as documented in the general practitioners' patient file and whether these reasons affected the time between medication discontinuation and death. METHODS: We performed a retrospective cohort study using the routine primary care database of the Julius General Practitioners' Network of the University Medical Centre Utrecht, a database consisting of routine care data from GPs from the city of Utrecht and its vicinity. Patients in the homecare setting with a life-limiting illness, diagnosed at least one year before death, were included. Descriptive analyses were used to describe the study population and the frequency of starting, using, and discontinuing medication and supplements in the last year of life. RESULTS: A total of 458 of 666 included patients (69%) used at least one preventive drug in the last year of life. Vitamins were used by 36% of the patients, followed with 35% using cholesterol-lowering medication, 24% using calcium supplements and 9% using bisphosphonates. Bisphosphonates were discontinued by 70% of the users, calcium supplements by 61%, vitamins by 56% and cholesterol-lowering medication by 48% of the users, with a median interval between day of discontinuation and death of 119, 60, 110 and, 65 days, respectively. The median time between medication or supplement discontinuation and death was longest in patients with side effects and who had medication reviews. CONCLUSION: Many patients in their last phase of life in the home-care setting use inappropriate medication and supplements. Timely medication review may contribute to optimise medication use in the last year of life.

Patient-Reported Outcomes During Checkpoint Inhibition: Insight into Symptom Burden in Daily Clinical Practice.

CONTEXT: While praised for inducing durable anti-tumour responses, immune checkpoint inhibitors (ICI) also cause immune-related adverse events (irAEs) that can vary in severity and affect health-related quality of life (HRQL). OBJECTIVES: This study was performed to provide insight into the course of symptoms and the influence of irAEs on HRQL measured with the treatment-specific Utrecht Symptom Diary Immunotherapy (USD-I). METHODS: In this observational cohort study, melanoma or non-small lung cancer (NSCLC) patients treated with PD(L)1-inhibitors between February 2016 and December 2018 were included. Data on symptoms, wellbeing and influence of side effects on HRQL were obtained using the patient-scored, treatment-specific USD-I, which was completed as part of routine care. Patients scored symptom intensity on a 0-10 numeric rating scale (NRS); NRS≥3 considered clinically relevant. RESULTS: A total of 162 melanoma (55%) or NSCLC (45%) patients completed 1493 USDs (median seven per patient). Most common patient-reported clinically relevant symptoms were: inactivity, fatigue, pain, cough and sleeping problems. Symptom prevalence decreased during treatment. Patients generally reported a low influence of side effects on HRQL. A higher number of clinically relevant symptoms at a certain time point correlated with poorer wellbeing. CONCLUSIONS: These data illustrate that ICI-treatment is generally well tolerated. However, especially the number of clinically relevant symptoms can impact patients wellbeing. Systematic use of an ICI-tailored PROM could create a window to discuss symptoms in a structured way which may promote personalized care during treatment.

Validation of the Dutch version of the Edmonton Symptom Assessment System.

BACKGROUND: The Utrecht Symptom Diary (USD) is a Dutch and adapted version of the Edmonton Symptom Assessment System, a patient-reported outcome measurement (PROM) tool to asses and monitor symptoms in cancer patients. This study analyses the validity and responsiveness of the USD and the cutoff points to determine the clinical significance of a symptom score. METHODS: Observational longitudinal cohort study including adult in- and outpatients treated in an academic hospital in the Netherlands who completed at least one USD as part of routine care (2012-2019). The distress thermometer and problem checklist (DT&PC) was used as a reference PROM. Content, construct and criterion validity, responsiveness, and cutoff points are shown with prevalences, area under receiver operating characteristic (ROC) curve, Chi-squared test, Wilcoxon signed-rank test, and positive and negative predictive values, respectively. RESULTS: A total of 3913 patients completed 22 400 USDs. Content validity was confirmed for all added USD items with prevalences of ≥22%. All USD items also present on the DT&PC demonstrated a good criterion validity (ROC >0.8). Construct validity was confirmed for the USD as a whole and for the items dry mouth, dysphagia and well-being (P < .0001). USD scores differed significantly for patients when improving or deteriorating on the DT&PC which confirmed responsiveness. Optimal cutoff points (3 or 4) differed per symptom. CONCLUSION: The USD is a valid 12-item PROM for the most prevalent symptoms in cancer patients, which has content, criterion, and construct validity, and detects clinically important changes over time, in both curative and palliative phase.

Predicting Anxiety in Hospitalized Cancer Patients.

CONTEXT: Anxiety in patients with cancer is highly prevalent; yet it remains underestimated and inadequately assessed. Little is known about predictors for anxiety in hospitalized patients with cancer. Insight in predictors should improve recognition and enable a targeted approach. OBJECTIVES: To determine the prevalence of anxiety and predictors for anxiety in hospitalized patients with cancer at different stages of disease. METHODS: A cross-sectional analysis of patients with cancer admitted to the Utrecht University Medical Center in 2015-2018 was conducted. The Utrecht Symptom Diary, an adapted Dutch version of the Edmonton Symptom Assessment System, was used to assess symptom burden on a numeric rating scale (0 = no symptom and 10 = worst possible symptom). Scores ≥4 were considered clinically relevant. All patients completed the Utrecht Symptom Diary as part of routine care. The first questionnaire after admission was selected. Using multivariable linear regression, the predictive value of potential predictors on anxiety was analyzed. RESULTS: In total, 2144 patients were included, of which 22% reported clinically relevant anxiety. The prevalence of anxiety was highest (36%) in patients receiving symptom-directed palliation only. In the total group, female gender, younger age, depressed mood, sleeping problems, dyspnea, and cancer of the head and neck were predictive of anxiety. Throughout all stages of disease, depressed mood was consistently the strongest predictor. CONCLUSION: We found a high prevalence of anxiety in hospitalized patients with cancer. It is recommended to explore anxiety in hospitalized patients with cancer, in particular when they experience depressed mood. Structural use of a symptom diary during hospitalization facilitates the recognition of anxiety and concurrent symptoms.

Genetic susceptibility to radiation-induced breast cancer after Hodgkin lymphoma.

Female Hodgkin lymphoma (HL) patients treated with chest radiotherapy (RT) have a very high risk of breast cancer. The contribution of genetic factors to this risk is unclear. We therefore examined 211 155 germline single-nucleotide polymorphisms (SNPs) for gene-radiation interaction on breast cancer risk in a case-only analysis including 327 breast cancer patients after chest RT for HL and 4671 first primary breast cancer patients. Nine SNPs showed statistically significant interaction with RT on breast cancer risk (false discovery rate, <20%), of which 1 SNP in the PVT1 oncogene attained the Bonferroni threshold for statistical significance. A polygenic risk score (PRS) composed of these SNPs (RT-interaction-PRS) and a previously published breast cancer PRS (BC-PRS) derived in the general population were evaluated in a case-control analysis comprising the 327 chest-irradiated HL patients with breast cancer and 491 chest-irradiated HL patients without breast cancer. Patients in the highest tertile of the RT-interaction-PRS had a 1.6-fold higher breast cancer risk than those in the lowest tertile. Remarkably, we observed a fourfold increased RT-induced breast cancer risk in the highest compared with the lowest decile of the BC-PRS. On a continuous scale, breast cancer risk increased 1.4-fold per standard deviation of the BC-PRS, similar to the effect size found in the general population. This study demonstrates that genetic factors influence breast cancer risk after chest RT for HL. Given the high absolute breast cancer risk in radiation-exposed women, these results can have important implications for the management of current HL survivors and future patients.

No clinical utility of KRAS variant rs61764370 for ovarian or breast cancer.

OBJECTIVE: Clinical genetic testing is commercially available for rs61764370, an inherited variant residing in a KRAS 3' UTR microRNA binding site, based on suggested associations with increased ovarian and breast cancer risk as well as with survival time. However, prior studies, emphasizing particular subgroups, were relatively small. Therefore, we comprehensively evaluated ovarian and breast cancer risks as well as clinical outcome associated with rs61764370. METHODS: Centralized genotyping and analysis were performed for 140,012 women enrolled in the Ovarian Cancer Association Consortium (15,357 ovarian cancer patients; 30,816 controls), the Breast Cancer Association Consortium (33,530 breast cancer patients; 37,640 controls), and the Consortium of Modifiers of BRCA1 and BRCA2 (14,765 BRCA1 and 7904 BRCA2 mutation carriers). RESULTS: We found no association with risk of ovarian cancer (OR=0.99, 95% CI 0.94-1.04, p=0.74) or breast cancer (OR=0.98, 95% CI 0.94-1.01, p=0.19) and results were consistent among mutation carriers (BRCA1, ovarian cancer HR=1.09, 95% CI 0.97-1.23, p=0.14, breast cancer HR=1.04, 95% CI 0.97-1.12, p=0.27; BRCA2, ovarian cancer HR=0.89, 95% CI 0.71-1.13, p=0.34, breast cancer HR=1.06, 95% CI 0.94-1.19, p=0.35). Null results were also obtained for associations with overall survival following ovarian cancer (HR=0.94, 95% CI 0.83-1.07, p=0.38), breast cancer (HR=0.96, 95% CI 0.87-1.06, p=0.38), and all other previously-reported associations. CONCLUSIONS: rs61764370 is not associated with risk of ovarian or breast cancer nor with clinical outcome for patients with these cancers. Therefore, genotyping this variant has no clinical utility related to the prediction or management of these cancers.

Candidate genetic modifiers for breast and ovarian cancer risk in BRCA1 and BRCA2 mutation carriers.

BACKGROUND: BRCA1 and BRCA2 mutation carriers are at substantially increased risk for developing breast and ovarian cancer. The incomplete penetrance coupled with the variable age at diagnosis in carriers of the same mutation suggests the existence of genetic and nongenetic modifying factors. In this study, we evaluated the putative role of variants in many candidate modifier genes. METHODS: Genotyping data from 15,252 BRCA1 and 8,211 BRCA2 mutation carriers, for known variants (n = 3,248) located within or around 445 candidate genes, were available through the iCOGS custom-designed array. Breast and ovarian cancer association analysis was performed within a retrospective cohort approach. RESULTS: The observed P values of association ranged between 0.005 and 1.000. None of the variants was significantly associated with breast or ovarian cancer risk in either BRCA1 or BRCA2 mutation carriers, after multiple testing adjustments. CONCLUSION: There is little evidence that any of the evaluated candidate variants act as modifiers of breast and/or ovarian cancer risk in BRCA1 or BRCA2 mutation carriers. IMPACT: Genome-wide association studies have been more successful at identifying genetic modifiers of BRCA1/2 penetrance than candidate gene studies.

Consent in psychiatric biobanks for pharmacogenetic research.

In psychiatric practice, pharmacogenetics has the potential to identify patients with an increased risk of unsatisfactory drug responses. Genotype-guided treatment adjustments may increase benefits and reduce harm in these patients; however, pharmacogenetic testing is not (yet) common practice and more pharmacogenetic research in psychiatric patients is warranted. An important precondition for this type of research is the establishment of biobanks. In this paper, we argue that, for the storage of samples in psychiatric biobanks, waiving of consent is not ethically justifiable since the risks cannot be considered minimal and the argument of impracticability does not apply. An opt-out consent procedure is also not justifiable, since it presumes competence while the decisional competence of psychiatric patients needs to be carefully evaluated. We state that an enhanced opt-in consent procedure is ethically necessary, i.e. a procedure that supports the patients' decision-making at the time when the patient is most competent. Nevertheless, such a procedure is not the traditional exhaustive informed consent procedure, since this is not feasible in the case of biobanking.

Potential of adaptive clinical trial designs in pharmacogenetic research.

Adaptive trial designs can be beneficial in pharmacogenetic research when prior uncertainty exists regarding the exact role and clinical relevance of genetic variability in drug response. This type of design enables us to learn about the effect of the genetic variability on drug response and to immediately use this information for the remainder of the study. For different types of adaptive trial designs, we discuss when and how the designs are suitable for pharmacogenetic research: adaptation of randomization, adaptation of patient enrollment and adaptive enrichment. To illustrate the potential benefits of an adaptive design over a fixed design, we simulated an adaptive trial based on the results of the IPASS trial. With a simple model we show that for this example an adaptive enrichment design would have led to a smaller trial, with less EGF receptor mutation-negative patients unnecessarily exposed to the drug, without compromising the α level or reducing power.

Pharmacogenetics in randomized controlled trials: considerations for trial design.

Pharmacogenetic analyses of clinical trials aim to either detect whether a subgroup of patients identified by genetic characteristics responds differently to the treatment or to verify whether a proposed genotype-guided treatment is beneficial over standard care. This article describes three different trial designs, differing in the timing of randomization and genotyping. Each design has its own advantages, and the objectives and conditions under which each one is most suited are discussed.

Hemoglobin variability in patients with chronic kidney disease in the Netherlands.

PURPOSE: Hemoglobin cycling has been reported in hemodialysis patients treated with erythropoiesis-stimulating agents (ESA) and is associated with increased mortality. Information on hemoglobin cycling in Europe is limited. We investigated hemoglobin variability in the Netherlands. Hemodialysis and peritoneal dialysis patients were studied and pre-dialysis patients were enrolled. METHODS: This observational retrospective study was executed in a Dutch dialysis center. We studied 157 patients from 2005 to 2007: 56 hemodialysis, 12 peritoneal dialysis and 29 pre-dialysis patients, all treated with ESA; and 60 pre-dialysis patients without ESA. Patients were divided on the basis of their pattern of hemoglobin fluctuation around a range of 11-12 g/dL. In dialysis patients, the amount of time that hemoglobin remained within that range was calculated. For all patients, the magnitude of hemoglobin fluctuations was assessed (i.e. the difference between hemoglobin maximum and minimum) and data on ESA dose changes and hospitalizations were collected. RESULTS: None of the ESA treated patients had hemoglobin levels stable within the target range over a one-year period. Pre-dialysis patients without ESA also showed variable hemoglobin levels. A stepwise decrease in the magnitudes of hemoglobin fluctuation was observed in the hemodialysis patients, peritoneal dialysis patients, pre-dialysis patients using ESA, and the pre-dialysis patients without ESA, respectively. CONCLUSION: In the Netherlands, hemoglobin variability is common in hemodialysis and peritoneal dialysis patients, but also in pre-dialysis patients. The results of this study warrant further research into the relationship between hemoglobin variability and clinical outcomes.