eduKateSG · Why Science?
Translate cancer biology into better evidence—connecting initiation, heterogeneity, tumour ecosystems, biomarkers, early detection, prevention, precision oncology, therapeutics, trials, data and patient outcomes
A fact-checked guide to NUS Centre for Cancer Research, its SCOUT and COMPASS themes, translational cancer research and safe student learning.
NUS states that N2CR was established in 2020 as a Translational Research Programme bringing together basic, translational and clinical scientists.
Its current official themes are SCOUT, focused on early events, precursor lesions and predictive biomarkers with emphasis on Asian-centric cancers, and COMPASS, focused on molecular profiling and precision oncology.
Did you know? Two tumours with the same organ label can contain different molecular changes and cell communities, which is one reason treatment responses vary.
N2CR and NUS control current research and programmes; qualified healthcare teams control diagnosis and treatment. This educational guide was source-checked 11 November 2026.
Chapter index
Chapters 1–9: cancer research, normal regulation, initiation, mutations, epigenetics, tumour microenvironment, immune interactions, heterogeneity, model systems
Chapters 10–17: biomarkers, early detection, prevention, molecular profiling, precision oncology, therapeutics, clinical trials, equity and evidence receipts
Cancer involves altered growth, survival, interaction and evolution across cells and tissues. Cancer research follows changes from normal regulation to altered cells, tissues and whole-person disease. NUS N2CR connects fundamental, translational and clinical researchers with a mission to detect, prevent and treat cancer. Chapter 1 keeps the evidence ladder visible so a molecular observation is not mistaken for a diagnostic test or treatment claim.
Connect molecular mechanisms to tissue context and person-level outcomes. A careful study records tumour type, stage, tissue handling, model system, assay, comparator, endpoint and analysis plan. Researchers use controls and independent cohorts because cancer is heterogeneous: cells from the same named disease can differ across people, sites and time. Reproducibility depends on preserving both biological context and technical provenance.
One mutation or hallmark does not explain every tumour. The clinical boundary is essential. An association is not necessarily a driver, activity in cells or animals does not establish benefit in people, and a biomarker may fail outside the cohort where it was discovered. Translation requires analytical validation, clinical validation, utility, safety, ethics and authorised trials.
Investigate “Define cancer as a system problem” with paper models or public, de-identified teaching datasets. Compare a control distribution with a fictional tumour-marker distribution and discuss overlap. Build a multiscale map from gene to patient. Never use classroom work to interpret a real person’s symptoms, genetic results, pathology or treatment.
Did you know? A tumour is an evolving ecosystem containing cancer cells, immune cells, blood vessels and supporting tissue. Treatment can change which cell populations survive. This helps learners see why one biopsy, one marker or one time point may not describe the complete future course.
Possible study routes include biology, chemistry, pharmacology, bioengineering, medicine, statistics, computing, pathology and public health. N2CR’s official pages describe current research themes, while programme and professional requirements can change. The responsible promise is a stronger evidence habit, never a guaranteed cure, admission or career.
Normal cells coordinate division, repair, differentiation and death through regulated networks. Cancer research follows changes from normal regulation to altered cells, tissues and whole-person disease. NUS N2CR connects fundamental, translational and clinical researchers with a mission to detect, prevent and treat cancer. Chapter 2 keeps the evidence ladder visible so a molecular observation is not mistaken for a diagnostic test or treatment claim.
Use matched controls and perturb one pathway at a time. A careful study records tumour type, stage, tissue handling, model system, assay, comparator, endpoint and analysis plan. Researchers use controls and independent cohorts because cancer is heterogeneous: cells from the same named disease can differ across people, sites and time. Reproducibility depends on preserving both biological context and technical provenance.
A difference from normal does not automatically drive cancer. The clinical boundary is essential. An association is not necessarily a driver, activity in cells or animals does not establish benefit in people, and a biomarker may fail outside the cohort where it was discovered. Translation requires analytical validation, clinical validation, utility, safety, ethics and authorised trials.
Investigate “Study normal regulation first” with paper models or public, de-identified teaching datasets. Compare a control distribution with a fictional tumour-marker distribution and discuss overlap. Draw a control loop for cell division. Never use classroom work to interpret a real person’s symptoms, genetic results, pathology or treatment.
Did you know? A tumour is an evolving ecosystem containing cancer cells, immune cells, blood vessels and supporting tissue. Treatment can change which cell populations survive. This helps learners see why one biopsy, one marker or one time point may not describe the complete future course.
Possible study routes include biology, chemistry, pharmacology, bioengineering, medicine, statistics, computing, pathology and public health. N2CR’s official pages describe current research themes, while programme and professional requirements can change. The responsible promise is a stronger evidence habit, never a guaranteed cure, admission or career.
Early lesions accumulate changes under tissue and exposure contexts. Cancer research follows changes from normal regulation to altered cells, tissues and whole-person disease. NUS N2CR connects fundamental, translational and clinical researchers with a mission to detect, prevent and treat cancer. Chapter 3 keeps the evidence ladder visible so a molecular observation is not mistaken for a diagnostic test or treatment claim.
Use longitudinal, lineage and comparative evidence. A careful study records tumour type, stage, tissue handling, model system, assay, comparator, endpoint and analysis plan. Researchers use controls and independent cohorts because cancer is heterogeneous: cells from the same named disease can differ across people, sites and time. Reproducibility depends on preserving both biological context and technical provenance.
An early alteration may be common without being sufficient for malignancy. The clinical boundary is essential. An association is not necessarily a driver, activity in cells or animals does not establish benefit in people, and a biomarker may fail outside the cohort where it was discovered. Translation requires analytical validation, clinical validation, utility, safety, ethics and authorised trials.
Investigate “Trace initiation and precursors” with paper models or public, de-identified teaching datasets. Compare a control distribution with a fictional tumour-marker distribution and discuss overlap. Order a fictional sequence of precursor events. Never use classroom work to interpret a real person’s symptoms, genetic results, pathology or treatment.
Did you know? A tumour is an evolving ecosystem containing cancer cells, immune cells, blood vessels and supporting tissue. Treatment can change which cell populations survive. This helps learners see why one biopsy, one marker or one time point may not describe the complete future course.
Possible study routes include biology, chemistry, pharmacology, bioengineering, medicine, statistics, computing, pathology and public health. N2CR’s official pages describe current research themes, while programme and professional requirements can change. The responsible promise is a stronger evidence habit, never a guaranteed cure, admission or career.
Somatic variants can be drivers, passengers or uncertain findings. Cancer research follows changes from normal regulation to altered cells, tissues and whole-person disease. NUS N2CR connects fundamental, translational and clinical researchers with a mission to detect, prevent and treat cancer. Chapter 4 keeps the evidence ladder visible so a molecular observation is not mistaken for a diagnostic test or treatment claim.
Combine recurrence, function and clinical association. A careful study records tumour type, stage, tissue handling, model system, assay, comparator, endpoint and analysis plan. Researchers use controls and independent cohorts because cancer is heterogeneous: cells from the same named disease can differ across people, sites and time. Reproducibility depends on preserving both biological context and technical provenance.
Frequency alone does not establish mechanism. The clinical boundary is essential. An association is not necessarily a driver, activity in cells or animals does not establish benefit in people, and a biomarker may fail outside the cohort where it was discovered. Translation requires analytical validation, clinical validation, utility, safety, ethics and authorised trials.
Investigate “Read mutations in context” with paper models or public, de-identified teaching datasets. Compare a control distribution with a fictional tumour-marker distribution and discuss overlap. Classify fictional variants by evidence type. Never use classroom work to interpret a real person’s symptoms, genetic results, pathology or treatment.
Did you know? A tumour is an evolving ecosystem containing cancer cells, immune cells, blood vessels and supporting tissue. Treatment can change which cell populations survive. This helps learners see why one biopsy, one marker or one time point may not describe the complete future course.
Possible study routes include biology, chemistry, pharmacology, bioengineering, medicine, statistics, computing, pathology and public health. N2CR’s official pages describe current research themes, while programme and professional requirements can change. The responsible promise is a stronger evidence habit, never a guaranteed cure, admission or career.
| Evidence move | What to record | What it cannot prove alone |
|---|---|---|
| Biology question | Population, variable, unit and time window | A definition does not establish a result |
| Comparison | Baseline, comparator and uncertainty | Association alone does not establish cause |
| Decision boundary | Owner, trigger, review and escalation | A model does not replace accountable judgement |
Chromatin and DNA regulation can alter gene activity without changing sequence. Cancer research follows changes from normal regulation to altered cells, tissues and whole-person disease. NUS N2CR connects fundamental, translational and clinical researchers with a mission to detect, prevent and treat cancer. Chapter 5 keeps the evidence ladder visible so a molecular observation is not mistaken for a diagnostic test or treatment claim.
Measure marks with cell-type and technical controls. A careful study records tumour type, stage, tissue handling, model system, assay, comparator, endpoint and analysis plan. Researchers use controls and independent cohorts because cancer is heterogeneous: cells from the same named disease can differ across people, sites and time. Reproducibility depends on preserving both biological context and technical provenance.
Association with expression does not prove causal direction. The clinical boundary is essential. An association is not necessarily a driver, activity in cells or animals does not establish benefit in people, and a biomarker may fail outside the cohort where it was discovered. Translation requires analytical validation, clinical validation, utility, safety, ethics and authorised trials.
Investigate “Include epigenetic change” with paper models or public, de-identified teaching datasets. Compare a control distribution with a fictional tumour-marker distribution and discuss overlap. Compare genetic and epigenetic evidence maps. Never use classroom work to interpret a real person’s symptoms, genetic results, pathology or treatment.
Did you know? A tumour is an evolving ecosystem containing cancer cells, immune cells, blood vessels and supporting tissue. Treatment can change which cell populations survive. This helps learners see why one biopsy, one marker or one time point may not describe the complete future course.
Possible study routes include biology, chemistry, pharmacology, bioengineering, medicine, statistics, computing, pathology and public health. N2CR’s official pages describe current research themes, while programme and professional requirements can change. The responsible promise is a stronger evidence habit, never a guaranteed cure, admission or career.
Immune, stromal and vascular cells interact with cancer cells and treatment. Cancer research follows changes from normal regulation to altered cells, tissues and whole-person disease. NUS N2CR connects fundamental, translational and clinical researchers with a mission to detect, prevent and treat cancer. Chapter 6 keeps the evidence ladder visible so a molecular observation is not mistaken for a diagnostic test or treatment claim.
Use spatial and functional measurements. A careful study records tumour type, stage, tissue handling, model system, assay, comparator, endpoint and analysis plan. Researchers use controls and independent cohorts because cancer is heterogeneous: cells from the same named disease can differ across people, sites and time. Reproducibility depends on preserving both biological context and technical provenance.
Bulk averages can hide opposing cell populations. The clinical boundary is essential. An association is not necessarily a driver, activity in cells or animals does not establish benefit in people, and a biomarker may fail outside the cohort where it was discovered. Translation requires analytical validation, clinical validation, utility, safety, ethics and authorised trials.
Investigate “Map the tumour microenvironment” with paper models or public, de-identified teaching datasets. Compare a control distribution with a fictional tumour-marker distribution and discuss overlap. Annotate a fictional tumour neighbourhood. Never use classroom work to interpret a real person’s symptoms, genetic results, pathology or treatment.
Did you know? A tumour is an evolving ecosystem containing cancer cells, immune cells, blood vessels and supporting tissue. Treatment can change which cell populations survive. This helps learners see why one biopsy, one marker or one time point may not describe the complete future course.
Possible study routes include biology, chemistry, pharmacology, bioengineering, medicine, statistics, computing, pathology and public health. N2CR’s official pages describe current research themes, while programme and professional requirements can change. The responsible promise is a stronger evidence habit, never a guaranteed cure, admission or career.
Tumours can be recognised, ignored or protected within immune contexts. Cancer research follows changes from normal regulation to altered cells, tissues and whole-person disease. NUS N2CR connects fundamental, translational and clinical researchers with a mission to detect, prevent and treat cancer. Chapter 7 keeps the evidence ladder visible so a molecular observation is not mistaken for a diagnostic test or treatment claim.
Measure cell state, location and function together. A careful study records tumour type, stage, tissue handling, model system, assay, comparator, endpoint and analysis plan. Researchers use controls and independent cohorts because cancer is heterogeneous: cells from the same named disease can differ across people, sites and time. Reproducibility depends on preserving both biological context and technical provenance.
Presence of immune cells does not prove effective anti-tumour activity. The clinical boundary is essential. An association is not necessarily a driver, activity in cells or animals does not establish benefit in people, and a biomarker may fail outside the cohort where it was discovered. Translation requires analytical validation, clinical validation, utility, safety, ethics and authorised trials.
Investigate “Understand immune interactions” with paper models or public, de-identified teaching datasets. Compare a control distribution with a fictional tumour-marker distribution and discuss overlap. Compare two fictional immune profiles. Never use classroom work to interpret a real person’s symptoms, genetic results, pathology or treatment.
Did you know? A tumour is an evolving ecosystem containing cancer cells, immune cells, blood vessels and supporting tissue. Treatment can change which cell populations survive. This helps learners see why one biopsy, one marker or one time point may not describe the complete future course.
Possible study routes include biology, chemistry, pharmacology, bioengineering, medicine, statistics, computing, pathology and public health. N2CR’s official pages describe current research themes, while programme and professional requirements can change. The responsible promise is a stronger evidence habit, never a guaranteed cure, admission or career.
Tumours vary within and between people and change under selection. Cancer research follows changes from normal regulation to altered cells, tissues and whole-person disease. NUS N2CR connects fundamental, translational and clinical researchers with a mission to detect, prevent and treat cancer. Chapter 8 keeps the evidence ladder visible so a molecular observation is not mistaken for a diagnostic test or treatment claim.
Sample across space and time where justified. A careful study records tumour type, stage, tissue handling, model system, assay, comparator, endpoint and analysis plan. Researchers use controls and independent cohorts because cancer is heterogeneous: cells from the same named disease can differ across people, sites and time. Reproducibility depends on preserving both biological context and technical provenance.
One biopsy may not capture every relevant clone. The clinical boundary is essential. An association is not necessarily a driver, activity in cells or animals does not establish benefit in people, and a biomarker may fail outside the cohort where it was discovered. Translation requires analytical validation, clinical validation, utility, safety, ethics and authorised trials.
Investigate “Track heterogeneity and evolution” with paper models or public, de-identified teaching datasets. Compare a control distribution with a fictional tumour-marker distribution and discuss overlap. Draw a branching evolution model. Never use classroom work to interpret a real person’s symptoms, genetic results, pathology or treatment.
Did you know? A tumour is an evolving ecosystem containing cancer cells, immune cells, blood vessels and supporting tissue. Treatment can change which cell populations survive. This helps learners see why one biopsy, one marker or one time point may not describe the complete future course.
Possible study routes include biology, chemistry, pharmacology, bioengineering, medicine, statistics, computing, pathology and public health. N2CR’s official pages describe current research themes, while programme and professional requirements can change. The responsible promise is a stronger evidence habit, never a guaranteed cure, admission or career.
| Evidence move | What to record | What it cannot prove alone |
|---|---|---|
| Measurement question | Population, variable, unit and time window | A definition does not establish a result |
| Comparison | Baseline, comparator and uncertainty | Association alone does not establish cause |
| Decision boundary | Owner, trigger, review and escalation | A model does not replace accountable judgement |
Cell lines, organoids, animals and computational models answer different questions. Cancer research follows changes from normal regulation to altered cells, tissues and whole-person disease. NUS N2CR connects fundamental, translational and clinical researchers with a mission to detect, prevent and treat cancer. Chapter 9 keeps the evidence ladder visible so a molecular observation is not mistaken for a diagnostic test or treatment claim.
Match model strengths and limits to the claim. A careful study records tumour type, stage, tissue handling, model system, assay, comparator, endpoint and analysis plan. Researchers use controls and independent cohorts because cancer is heterogeneous: cells from the same named disease can differ across people, sites and time. Reproducibility depends on preserving both biological context and technical provenance.
A model response is not a patient outcome. The clinical boundary is essential. An association is not necessarily a driver, activity in cells or animals does not establish benefit in people, and a biomarker may fail outside the cohort where it was discovered. Translation requires analytical validation, clinical validation, utility, safety, ethics and authorised trials.
Investigate “Choose model systems” with paper models or public, de-identified teaching datasets. Compare a control distribution with a fictional tumour-marker distribution and discuss overlap. Build a model-selection matrix. Never use classroom work to interpret a real person’s symptoms, genetic results, pathology or treatment.
Did you know? A tumour is an evolving ecosystem containing cancer cells, immune cells, blood vessels and supporting tissue. Treatment can change which cell populations survive. This helps learners see why one biopsy, one marker or one time point may not describe the complete future course.
Possible study routes include biology, chemistry, pharmacology, bioengineering, medicine, statistics, computing, pathology and public health. N2CR’s official pages describe current research themes, while programme and professional requirements can change. The responsible promise is a stronger evidence habit, never a guaranteed cure, admission or career.
Biomarkers may indicate risk, diagnosis, prognosis, prediction or monitoring. Cancer research follows changes from normal regulation to altered cells, tissues and whole-person disease. NUS N2CR connects fundamental, translational and clinical researchers with a mission to detect, prevent and treat cancer. Chapter 10 keeps the evidence ladder visible so a molecular observation is not mistaken for a diagnostic test or treatment claim.
Separate discovery, analytical validation, clinical validation and utility. A careful study records tumour type, stage, tissue handling, model system, assay, comparator, endpoint and analysis plan. Researchers use controls and independent cohorts because cancer is heterogeneous: cells from the same named disease can differ across people, sites and time. Reproducibility depends on preserving both biological context and technical provenance.
A statistically significant marker may be too overlapping for decisions. The clinical boundary is essential. An association is not necessarily a driver, activity in cells or animals does not establish benefit in people, and a biomarker may fail outside the cohort where it was discovered. Translation requires analytical validation, clinical validation, utility, safety, ethics and authorised trials.
Investigate “Discover biomarkers” with paper models or public, de-identified teaching datasets. Compare a control distribution with a fictional tumour-marker distribution and discuss overlap. Plot fictional distributions and choose a threshold. Never use classroom work to interpret a real person’s symptoms, genetic results, pathology or treatment.
Did you know? A tumour is an evolving ecosystem containing cancer cells, immune cells, blood vessels and supporting tissue. Treatment can change which cell populations survive. This helps learners see why one biopsy, one marker or one time point may not describe the complete future course.
Possible study routes include biology, chemistry, pharmacology, bioengineering, medicine, statistics, computing, pathology and public health. N2CR’s official pages describe current research themes, while programme and professional requirements can change. The responsible promise is a stronger evidence habit, never a guaranteed cure, admission or career.
Early detection seeks useful signals before harmful disease advances. Cancer research follows changes from normal regulation to altered cells, tissues and whole-person disease. NUS N2CR connects fundamental, translational and clinical researchers with a mission to detect, prevent and treat cancer. Chapter 11 keeps the evidence ladder visible so a molecular observation is not mistaken for a diagnostic test or treatment claim.
Evaluate sensitivity, specificity, prevalence and downstream work-up. A careful study records tumour type, stage, tissue handling, model system, assay, comparator, endpoint and analysis plan. Researchers use controls and independent cohorts because cancer is heterogeneous: cells from the same named disease can differ across people, sites and time. Reproducibility depends on preserving both biological context and technical provenance.
More detection can include overdiagnosis and false positives. The clinical boundary is essential. An association is not necessarily a driver, activity in cells or animals does not establish benefit in people, and a biomarker may fail outside the cohort where it was discovered. Translation requires analytical validation, clinical validation, utility, safety, ethics and authorised trials.
Investigate “Improve early detection” with paper models or public, de-identified teaching datasets. Compare a control distribution with a fictional tumour-marker distribution and discuss overlap. Compare screening outcomes in high- and low-risk groups. Never use classroom work to interpret a real person’s symptoms, genetic results, pathology or treatment.
Did you know? A tumour is an evolving ecosystem containing cancer cells, immune cells, blood vessels and supporting tissue. Treatment can change which cell populations survive. This helps learners see why one biopsy, one marker or one time point may not describe the complete future course.
Possible study routes include biology, chemistry, pharmacology, bioengineering, medicine, statistics, computing, pathology and public health. N2CR’s official pages describe current research themes, while programme and professional requirements can change. The responsible promise is a stronger evidence habit, never a guaranteed cure, admission or career.
Prevention can target exposures, infections, inherited risk and precursor conditions. Cancer research follows changes from normal regulation to altered cells, tissues and whole-person disease. NUS N2CR connects fundamental, translational and clinical researchers with a mission to detect, prevent and treat cancer. Chapter 12 keeps the evidence ladder visible so a molecular observation is not mistaken for a diagnostic test or treatment claim.
Measure benefits, harms, uptake and equity over suitable periods. A careful study records tumour type, stage, tissue handling, model system, assay, comparator, endpoint and analysis plan. Researchers use controls and independent cohorts because cancer is heterogeneous: cells from the same named disease can differ across people, sites and time. Reproducibility depends on preserving both biological context and technical provenance.
Risk reduction is not a guarantee for an individual. The clinical boundary is essential. An association is not necessarily a driver, activity in cells or animals does not establish benefit in people, and a biomarker may fail outside the cohort where it was discovered. Translation requires analytical validation, clinical validation, utility, safety, ethics and authorised trials.
Investigate “Study prevention” with paper models or public, de-identified teaching datasets. Compare a control distribution with a fictional tumour-marker distribution and discuss overlap. Build a prevention evidence ladder. Never use classroom work to interpret a real person’s symptoms, genetic results, pathology or treatment.
Did you know? A tumour is an evolving ecosystem containing cancer cells, immune cells, blood vessels and supporting tissue. Treatment can change which cell populations survive. This helps learners see why one biopsy, one marker or one time point may not describe the complete future course.
Possible study routes include biology, chemistry, pharmacology, bioengineering, medicine, statistics, computing, pathology and public health. N2CR’s official pages describe current research themes, while programme and professional requirements can change. The responsible promise is a stronger evidence habit, never a guaranteed cure, admission or career.
| Evidence move | What to record | What it cannot prove alone |
|---|---|---|
| Translation question | Population, variable, unit and time window | A definition does not establish a result |
| Comparison | Baseline, comparator and uncertainty | Association alone does not establish cause |
| Decision boundary | Owner, trigger, review and escalation | A model does not replace accountable judgement |
Profiling characterises DNA, RNA, proteins or cells to refine biology and selection. Cancer research follows changes from normal regulation to altered cells, tissues and whole-person disease. NUS N2CR connects fundamental, translational and clinical researchers with a mission to detect, prevent and treat cancer. Chapter 13 keeps the evidence ladder visible so a molecular observation is not mistaken for a diagnostic test or treatment claim.
Preserve specimen quality and validate assays. A careful study records tumour type, stage, tissue handling, model system, assay, comparator, endpoint and analysis plan. Researchers use controls and independent cohorts because cancer is heterogeneous: cells from the same named disease can differ across people, sites and time. Reproducibility depends on preserving both biological context and technical provenance.
A molecular category does not automatically reveal an effective therapy. The clinical boundary is essential. An association is not necessarily a driver, activity in cells or animals does not establish benefit in people, and a biomarker may fail outside the cohort where it was discovered. Translation requires analytical validation, clinical validation, utility, safety, ethics and authorised trials.
Investigate “Use molecular profiling” with paper models or public, de-identified teaching datasets. Compare a control distribution with a fictional tumour-marker distribution and discuss overlap. Create a sample-to-decision workflow. Never use classroom work to interpret a real person’s symptoms, genetic results, pathology or treatment.
Did you know? A tumour is an evolving ecosystem containing cancer cells, immune cells, blood vessels and supporting tissue. Treatment can change which cell populations survive. This helps learners see why one biopsy, one marker or one time point may not describe the complete future course.
Possible study routes include biology, chemistry, pharmacology, bioengineering, medicine, statistics, computing, pathology and public health. N2CR’s official pages describe current research themes, while programme and professional requirements can change. The responsible promise is a stronger evidence habit, never a guaranteed cure, admission or career.
Precision oncology matches evidence about tumour and person to a treatment strategy. Cancer research follows changes from normal regulation to altered cells, tissues and whole-person disease. NUS N2CR connects fundamental, translational and clinical researchers with a mission to detect, prevent and treat cancer. Chapter 14 keeps the evidence ladder visible so a molecular observation is not mistaken for a diagnostic test or treatment claim.
Compare biomarker-defined groups in suitable trials. A careful study records tumour type, stage, tissue handling, model system, assay, comparator, endpoint and analysis plan. Researchers use controls and independent cohorts because cancer is heterogeneous: cells from the same named disease can differ across people, sites and time. Reproducibility depends on preserving both biological context and technical provenance.
Personalised does not mean unique or certain. The clinical boundary is essential. An association is not necessarily a driver, activity in cells or animals does not establish benefit in people, and a biomarker may fail outside the cohort where it was discovered. Translation requires analytical validation, clinical validation, utility, safety, ethics and authorised trials.
Investigate “Develop precision oncology” with paper models or public, de-identified teaching datasets. Compare a control distribution with a fictional tumour-marker distribution and discuss overlap. Design a fictional stratified trial diagram. Never use classroom work to interpret a real person’s symptoms, genetic results, pathology or treatment.
Did you know? A tumour is an evolving ecosystem containing cancer cells, immune cells, blood vessels and supporting tissue. Treatment can change which cell populations survive. This helps learners see why one biopsy, one marker or one time point may not describe the complete future course.
Possible study routes include biology, chemistry, pharmacology, bioengineering, medicine, statistics, computing, pathology and public health. N2CR’s official pages describe current research themes, while programme and professional requirements can change. The responsible promise is a stronger evidence habit, never a guaranteed cure, admission or career.
Therapies may target signalling, DNA repair, immunity or tumour dependencies. Cancer research follows changes from normal regulation to altered cells, tissues and whole-person disease. NUS N2CR connects fundamental, translational and clinical researchers with a mission to detect, prevent and treat cancer. Chapter 15 keeps the evidence ladder visible so a molecular observation is not mistaken for a diagnostic test or treatment claim.
Use dose-response, selectivity and resistance studies. A careful study records tumour type, stage, tissue handling, model system, assay, comparator, endpoint and analysis plan. Researchers use controls and independent cohorts because cancer is heterogeneous: cells from the same named disease can differ across people, sites and time. Reproducibility depends on preserving both biological context and technical provenance.
Cell killing in vitro does not establish safe clinical benefit. The clinical boundary is essential. An association is not necessarily a driver, activity in cells or animals does not establish benefit in people, and a biomarker may fail outside the cohort where it was discovered. Translation requires analytical validation, clinical validation, utility, safety, ethics and authorised trials.
Investigate “Test therapeutic mechanisms” with paper models or public, de-identified teaching datasets. Compare a control distribution with a fictional tumour-marker distribution and discuss overlap. Map preclinical evidence gates. Never use classroom work to interpret a real person’s symptoms, genetic results, pathology or treatment.
Did you know? A tumour is an evolving ecosystem containing cancer cells, immune cells, blood vessels and supporting tissue. Treatment can change which cell populations survive. This helps learners see why one biopsy, one marker or one time point may not describe the complete future course.
Possible study routes include biology, chemistry, pharmacology, bioengineering, medicine, statistics, computing, pathology and public health. N2CR’s official pages describe current research themes, while programme and professional requirements can change. The responsible promise is a stronger evidence habit, never a guaranteed cure, admission or career.
Trials evaluate safety, dose, efficacy and comparative benefit under protocols. Cancer research follows changes from normal regulation to altered cells, tissues and whole-person disease. NUS N2CR connects fundamental, translational and clinical researchers with a mission to detect, prevent and treat cancer. Chapter 16 keeps the evidence ladder visible so a molecular observation is not mistaken for a diagnostic test or treatment claim.
Pre-register endpoints and monitor adverse events. A careful study records tumour type, stage, tissue handling, model system, assay, comparator, endpoint and analysis plan. Researchers use controls and independent cohorts because cancer is heterogeneous: cells from the same named disease can differ across people, sites and time. Reproducibility depends on preserving both biological context and technical provenance.
A phase result has a defined scope and may not generalise. The clinical boundary is essential. An association is not necessarily a driver, activity in cells or animals does not establish benefit in people, and a biomarker may fail outside the cohort where it was discovered. Translation requires analytical validation, clinical validation, utility, safety, ethics and authorised trials.
Investigate “Run clinical trials responsibly” with paper models or public, de-identified teaching datasets. Compare a control distribution with a fictional tumour-marker distribution and discuss overlap. Match trial phases to their main questions. Never use classroom work to interpret a real person’s symptoms, genetic results, pathology or treatment.
Did you know? A tumour is an evolving ecosystem containing cancer cells, immune cells, blood vessels and supporting tissue. Treatment can change which cell populations survive. This helps learners see why one biopsy, one marker or one time point may not describe the complete future course.
Possible study routes include biology, chemistry, pharmacology, bioengineering, medicine, statistics, computing, pathology and public health. N2CR’s official pages describe current research themes, while programme and professional requirements can change. The responsible promise is a stronger evidence habit, never a guaranteed cure, admission or career.
The receipt links disease context, sample, model, assay, statistics, translation stage and limits. Cancer research follows changes from normal regulation to altered cells, tissues and whole-person disease. NUS N2CR connects fundamental, translational and clinical researchers with a mission to detect, prevent and treat cancer. Chapter 17 keeps the evidence ladder visible so a molecular observation is not mistaken for a diagnostic test or treatment claim.
Archive consent, provenance, versions and negative results. A careful study records tumour type, stage, tissue handling, model system, assay, comparator, endpoint and analysis plan. Researchers use controls and independent cohorts because cancer is heterogeneous: cells from the same named disease can differ across people, sites and time. Reproducibility depends on preserving both biological context and technical provenance.
A complete receipt does not predict an individual course. The clinical boundary is essential. An association is not necessarily a driver, activity in cells or animals does not establish benefit in people, and a biomarker may fail outside the cohort where it was discovered. Translation requires analytical validation, clinical validation, utility, safety, ethics and authorised trials.
Investigate “Keep a cancer evidence receipt” with paper models or public, de-identified teaching datasets. Compare a control distribution with a fictional tumour-marker distribution and discuss overlap. Assemble a one-page research receipt. Never use classroom work to interpret a real person’s symptoms, genetic results, pathology or treatment.
Did you know? A tumour is an evolving ecosystem containing cancer cells, immune cells, blood vessels and supporting tissue. Treatment can change which cell populations survive. This helps learners see why one biopsy, one marker or one time point may not describe the complete future course.
Possible study routes include biology, chemistry, pharmacology, bioengineering, medicine, statistics, computing, pathology and public health. N2CR’s official pages describe current research themes, while programme and professional requirements can change. The responsible promise is a stronger evidence habit, never a guaranteed cure, admission or career.
| Evidence move | What to record | What it cannot prove alone |
|---|---|---|
| Cancer research question | Population, variable, unit and time window | A definition does not establish a result |
| Comparison | Baseline, comparator and uncertainty | Association alone does not establish cause |
| Decision boundary | Owner, trigger, review and escalation | A model does not replace accountable judgement |
Contents · Previous chapter · Continue in the Science Learning Hub
Primary sources and verification date
Source status for Why Science? | NUS Centre for Cancer Research was checked 11 November 2026. Research priorities, programmes, recruitment, facilities and participation routes can change; use the linked official owner for the current edition.
- NUS Centre for Cancer Research — Official centre, current themes and programme news.
- N2CR overview — Official vision, mission and translational role.
- N2CR SCOUT theme — Official early-cancer and Asian-centric research theme.
- N2CR COMPASS theme — Official precision-oncology research theme.
Continue the evidence journey
N2CR science becomes useful when learners can trace a cancer claim from mechanism through models and validation to a carefully governed question about patient benefit. Families considering science tuition or a science tutor can use the same evidence questions: what will the learner practise, what feedback will they receive, and how will progress be shown?