Colorectal studies run on scheduled imaging, biomarker-defined cohorts, long chemotherapy cycles and a steady stream of graded toxicities. Capture holds the tumor assessments, treatment cycles, CTCAE events and participant diaries in one study with one audit trail. Build it in the free sandbox, no credit card.
Free sandbox · No credit card · 21 CFR Part 11 aligned
| Assessment | Screen | C1 | C2 | C4 | C5 (wk 8) | C8 | C9 (wk 16) |
|---|---|---|---|---|---|---|---|
| CT/MRI tumor assessment (RECIST 1.1) | |||||||
| Treatment administration and dose | |||||||
| CTCAE adverse events | |||||||
| CEA and safety labs | |||||||
| Neuropathy diary (participant phone) | D | D | D | D | D | D |
x = done at this visit, D = diary between visits
What a colorectal trial needs from its EDC
The study data
Colorectal cancer is several protocol families under one name. Adjuvant colon cancer studies follow resected, node-positive or high-risk participants for recurrence, so the endpoint is disease-free survival and the main data burden is a long follow-up with scans, CEA and recurrence reporting. Metastatic studies treat measurable disease with doublet or triplet chemotherapy plus a targeted agent or immunotherapy, and the endpoints are progression-free survival, overall survival and response rate. Rectal studies add a locoregional dimension: chemoradiotherapy, surgical timing and pathological response. Each family has a different visit pattern, but they share a core of imaging, treatment exposure, biomarkers and toxicity.
The imaging data deserves the most care. Under RECIST 1.1, progression is a 20 percent increase in the sum of target-lesion diameters from the smallest sum on study, with an absolute increase of at least 5 mm, or unequivocal new lesions. In a registered metastatic colorectal protocol that uses blinded independent central review, the site assessment and the central assessment are both data, and they can disagree. Capture them as separate variables with their own dates, so the analysis can use whichever the protocol designates and the disagreement rate is visible. Liver-dominant disease, lung nodules and nodal lesions each need the lesion location recorded, because lesion-level data is what makes a re-read possible.
The second hard problem is dates. Progression-free survival is the time from randomization to the earliest documented progression or death. If the scan date, the date the radiologist read it and the date the site recorded it are all different fields, a monitor can verify the one that matters. If they are one free-text box, the PFS curve inherits the ambiguity. The same applies to the date of last contact for censoring, which is often the least-maintained date in a study.
Metastatic colorectal protocols commonly split by RAS and BRAF mutation status and by microsatellite instability or mismatch-repair deficiency, because these predict which therapies are relevant. Circulating tumor DNA is also appearing as a stratification or monitoring tool, including in ctDNA-guided adjuvant designs. Record the result, the assay, the sample date and the laboratory as typed fields next to the eligibility form, and add an edit check that stops an eligibility date being entered before the result date. Where a sample goes to a central laboratory, the lab results eCRF template gives the repeating-row structure.
Target lesions
Scan date
TUDTCLiver segment VII, longest diameter
TULDIAMRight paratracheal node, short axis
TULDIAMSum of diameters
TUSUMBaseline sum 58 mm
42 mm CalculatedNew lesions
TUNEWOverall response, site
RSORRESProtocol to build
| Protocol element | What the data looks like | Where it lives in Capture |
|---|---|---|
| Eligibility and biomarkers | Stage, resection status, RAS/BRAF/MSI result, performance status | Screening form with edit checks; ECOG form template |
| Tumor assessment | Target and non-target lesions, new lesions, overall response, per scan | Visit eCRF form with repeating lesion rows and a calculated sum |
| Treatment exposure | Cycle date, dose given, reductions, delays, reason | Repeating cycle form; BSA calculated from height and weight |
| Survival and follow-up | Progression date, death date, date last known alive, subsequent therapy | Follow-up form with date edit checks |
| Toxicity | AE term, CTCAE grade, seriousness, causality, action | Adverse event form and SAE report form |
| Patient-reported outcomes | Neuropathy and quality-of-life questionnaires | ePRO tasks on the phone; EORTC QLQ-C30 form structure, PRO-CTCAE |
| Blinded and randomized arms | Allocation, stratification (e.g. by biomarker or site) | Randomization in the same platform; blinded roles never receive arm values |
Questionnaire wording is yours to supply. Licensed instruments stay with your study documents; Capture does not supply or verify instrument licences, and licence controls exist only on participant questionnaires.
Upload the protocol, let the AI draft visits and forms for review, then enter practice participants through two cycles. Free sandbox, no credit card.
Safety reporting
Colorectal regimens produce a recognisable toxicity profile, and the protocol usually names which events must be graded at every cycle. Fluoropyrimidines bring diarrhea, mucositis and hand-foot syndrome. Oxaliplatin brings acute cold-triggered symptoms and a cumulative peripheral sensory neuropathy that can persist after treatment stops. Irinotecan brings diarrhea and neutropenia. Anti-EGFR antibodies bring skin rash and low magnesium, bevacizumab brings hypertension and bleeding, and immune checkpoint inhibitors in mismatch-repair-deficient disease bring immune-related events. Every one of these is graded on the CTCAE scale of 1 to 5, and the grade is only one of several answers the form needs.
Keep severity, seriousness and causality as separate fields. A grade 3 neutropenia is severe but not necessarily serious; a grade 2 event that causes hospitalization is serious. The CTCAE grade lookup shows the published grade criteria for common terms in both v5.0 and v6.0, which helps sites and data managers check a grade against the criteria; the investigator still assigns it. Serious events follow the SAE report form path, and the SAE deadline calculator helps teams plan reporting clocks. The wider workflow is described on the adverse event reporting software page.
Because neuropathy accumulates, clinic-only grading tends to lag. A short diary between visits, built from your own questions or the structure of PRO-CTCAE, shows the trend, and an edit check can raise a query when a reported symptom is worse than the grade recorded at the visit. Dose-modification rules in the protocol then have two data sources to work from. Treat the diary as supporting data for the investigator, not as a replacement for clinical assessment.
Cannot submit yet
Causality is required because the event is serious.
Try it
The same lookup that sits on its own page, here for checking terms such as diarrhea, neutrophil count decreased or peripheral sensory neuropathy.
| Grade 1 | <LLN - 1500/mm3; <LLN - 1.5 x 10e9 /L |
|---|---|
| Grade 2 | <1500 - 1000/mm3; <1.5 - 1.0 x 10e9 /L |
| Grade 3 | <1000 - 500/mm3; <1.0 - 0.5 x 10e9 /L |
| Grade 4 | <500/mm3; <0.5 x 10e9 /L |
| Grade 5 | – (grade not available) |
Source: NCI CTCAE v5.0, published 27 November 2017 (ctcae-v5.0.xlsx, dctd.cancer.gov).
*Instrumental ADL: preparing meals, shopping for groceries or clothes, using the telephone, managing money, etc. **Self care ADL: bathing, dressing and undressing, feeding self, using the toilet, taking medications, and not bedridden. A semicolon means “or” within a grade; a dash means the grade is not available for that term. LLN and ULN are the lower and upper limits of normal.
Reference only: 34 common terms, not the full CTCAE, and not validated software. Use the version your protocol names. Grading, seriousness and causality are the investigator's clinical judgement; always check the official NCI document.
Build sequence
A realistic order of work for a sponsor or investigator-initiated team.
Upload the protocol (PDF, DOCX or DOC) and let the AI study builder propose the schedule and forms. Nothing is saved until a person reviews it.
Repeating lesion rows, a calculated sum, new-lesion and overall-response fields, and a separate form if a central read is planned.
Range checks on lesion size and labs, date logic across scan, progression and death, and visit windows around each scan.
Neuropathy and quality-of-life questionnaires with reminders and completion windows, opened from a link on the phone.
Run practice participants through progression and death, check the exports, approve the forms and start the paid live phase when real participants enroll.
Follow-up and scale
Colorectal studies are long. Adjuvant designs follow participants for years, and metastatic studies continue survival follow-up after treatment stops. Plan a lean follow-up form that records survival status, date last known alive, subsequent therapy and any recurrence, and schedule it with a visit window so lapses are visible. The compliance view shows by study, country and site which participants are overdue.
Multi-center work is normal. Capture supports QR-code enrollment, site-level participant numbering, a separate site coordinator portal and by-site filtering and exports; see multi-site clinical trial management. For the analysis, CSV and Excel exports include an automatic data dictionary, and CDISC SDTM export produces SAS XPT datasets with Define-XML; the CDISC SDTM export page describes it. If you are sizing a survival endpoint, the survival sample size calculator gives a first estimate.
This page covers colorectal specifics. For dose escalation, BSA dosing and blinded oncology arms in general, see EDC for oncology clinical trials, and for multi-site pivotal designs, EDC for Phase 3 trials.
Before first participant
RECIST 1.1 or the protocol variant, with who reads the scan and whether a central read applies.
Which date is progression, which is censoring, which is last known alive.
Assay, sample date, laboratory and result for every stratification factor.
Which events are graded every cycle and which trigger a dose rule.
Questionnaire terms confirmed with their rights holders and kept in the study documents.
Wide CSV with data dictionary opened in the statistics package.
Yes. A tumor assessment is built as a site eCRF form with repeating rows for lesions, a calculated sum of diameters and fields for new lesions and overall response. The response rules themselves are applied by the investigator or the central reader, not by Capture.
Yes. Build them as separate forms or variables with their own dates, so the analysis can use the designated read and the disagreement rate is visible.
As a field on the adverse event form, next to separate fields for seriousness, relationship and action. The CTCAE grade lookup tool shows published grade criteria for common terms in v5.0 and v6.0. Assigning the grade stays with the investigator.
Yes. A short questionnaire opens from a secure link in the phone browser, with no app, reminders and a completion window you control. An edit check can prompt a query when the report conflicts with the grade recorded at the visit.
Yes. The forms and schedule are built from your protocol, so disease-free survival follow-up and progression-based endpoints use the same building blocks with different visit patterns.
Yes. Capture is used from Phase 1 through Phase 3, with site-level numbering, a site coordinator portal and by-site exports for multi-center work.
Capture provides Part 11-aligned controls: a field-level audit trail, electronic signatures and role-based access. Compliance is shared with the sponsor’s own validated use of the system. This is not legal advice.
Keep exploring
Free sandbox with every feature. No credit card, and you pay only when you go live with real participants.