Skip to main content

How does a Center identify and select patients (outliers and controls)?

Category: Study Design

Labels: framework, patient identification

Creation date and Author: Maria G., 23 June 2025 

Last Update and Author: Astrid C., July 31st 2025

Related questions: 

  • How to choose the matching control?
  • If a patient is treated with Carboplatin or Cisplatin, does his control need to have exactly the same type of platinum treatment?
  • How do you select control patients?
  • What are control patients? How are they included in the study?
  • How does a center identify and select cases and controls?
  • How often does the PI need to meet with the patients?
  • Can a patient who participates / has participated in several studies be included in the ROSALIND study?
  • What if deceased patients cannot be included?  
  • Why is enrolling deceased patients essential to the ROSALIND study? 
 
 
Cure51 provides the Principal Investigator (PI) with the study protocol, which outlines in detail the inclusion criteria for both outlier patients and their matched controls. The ROSALIND study is indeed a 1:1 study: for each outlier, a matched control is required. 
In addition to the protocol, we also share several supporting documents to facilitate patient identification, including:
  • Screening tool
  • Study initiation questionnaire
  • Rosalind controls (a detailed guide on the control matching criteria)It is entirely up to the PI to determine the most suitable approach for identifying eligible patients, based on what works best for their center. 
 
Typically, the process includes the following steps:

  • Initial Screening: Centers conduct a search of their patient database to identify potential outlier patients who meet the inclusion criteria.

  • Medical Record Review: Once a preliminary list is compiled, a more in-depth review of the medical records is performed to confirm eligibility. 

  • Control Matching: After the outlier patients are confirmed, the center proceeds to identify appropriately matched control patients.

  • Sample Availability: The final step is to confirm the availability of biological samples for both outlier and control patients.

 

NB: 

The measures for recruiting study participants must be ensured to comply with the locally applicable legal requirements.

Identifying information of the study subjects is not permitted to be accessed by Cure51. Instead, the Center is relied upon to identify the participants, provide their de-identified information and maintain the appropriate relationship with them to collect their consent where applicable.

 
 

Q&A

  • If a patient is treated with Carboplatin or Cisplatin, does his control need to have exactly the same type of platinum treatment?

A cisplatin/cisplatin or carboplatin/carboplatin match is preferred.

However, it is possible to make a match between a patient on carboplatin and a control on cisplatin, or vice versa.

  • What are control patients? How are they included in the study?

Control patients are paired with outliers: they have baseline characteristics and care journey as similar as possible to tat of an outlier. They are mandatory for our analyses. 

Inclusion Criteria

  1. Adult patient: patient over 18 years of age at diagnosis;

  2. Diagnosis of mPDAC, ES-SCLC or GBM IDH-wt;

  3. Paired to the long-term survivors on age, stage, and other prognostic factors for each specific cohort:

  4. Death on median overall survival as reported in pivotal clinical trials in the specific type diseases

  • How often does the PI need to meet with the patients? 

As this is a retrospective study, no regular contact with study participants is considered necessary beyond providing information or obtaining consent.

  • Can a patient who participates / has participated in several studies be included in the ROSALIND study? 

Participation of patients in other studies simultaneously with the ROSALIND study is permitted, provided that the inclusion criteria are fulfilled and sufficient sample material is available.

  • What if deceased patients cannot be included? 

If unfortunately, deceased patients cannot be included for regulatory purposes, the identified alive long-term survivors (outliers may also be deceased) may still be included in the study. 

However, matched control patients would then need to be sourced from other participating sites, which may introduce additional logistical challenges.

  • Why is enrolling deceased patients essential to the ROSALIND study? 

The inclusion of deceased control patients is an essential component of the ROSALIND study, as each long-term survivor requires a matched control patient. These control patients typically correspond to individuals who experienced median overall survival for the specific disease type and are, therefore, deceased.

The possibility of enrolling deceased patients would also substantially enhance the overall capacity to recruit long-term survivors, extending beyond those who are currently living.

In cases where deceased patients cannot be provided by a given site, the identified long-term survivors may still be included in the study. However, matched control patients would then need to be sourced from other participating sites, which may introduce additional logistical challenges.

Was this article helpful?