Feb 11, 2026

Survivorship after CANcer and its LAte effects (SCALA) Protocol v 1 (2025.07.09) V.2

This  protocol  is a draft, published without a DOI.
  • Pauline Boucheron1,
  • Valerie McCormack1
  • 1International Agency for Research on Cancer
  • Valerie McCormack: SCALA PI
Pauline Boucheron
Icon indicating open access to content
QR code linking to this content
Protocol Citation: Pauline Boucheron, Valerie McCormack 2026. Survivorship after CANcer and its LAte effects (SCALA) Protocol v 1 (2025.07.09). protocols.io https://dx.doi.org/Version created by Pauline Boucheron
License: This is an open access  protocol  distributed under the terms of the  Creative Commons Attribution License,  which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: Working
We use this protocol and it's working
Created: February 10, 2026
Last Modified: February 11, 2026
Protocol  Integer ID: 242982
Keywords: important aspect of cancer survivorship, survivorship after cancer, cancer survivor, cancer survivorship, survivors of various cancer type, social effects of cancer, term effects of cancer, tertiary prevention of cancer, major consequences of cancer, survivorship duration, cancer, survivorship, information on cancer diagnosis, cancer diagnosis, qol outcomes between prevalent c, comprehensive mapping of cancer, various cancer type, quality of life, cs beyond the initial treatment phase, survivor, large nationwide french cohort, baseline data from the constances cohort, scala study, qol outcome
Funders Acknowledgements:
Institut National du Cancer
Grant ID: INCA_19680
Disclaimer
Where authors are identified as personnel of the International Agency for Research on Cancer/World Health Organization, the authors alone are responsible for the views expressed and they do not necessarily represent the decisions, policy or views of the International Agency for Research on Cancer /World Health Organization.
Abstract
Background: Quality of life (QOL) is an important aspect of cancer survivorship encompassing physical, psychological and social well-being, which remains understudied while often altered in cancer survivors (CS) compared with individuals without a history of cancer. Improved knowledge of the profile and determinants of late and long-term effects of cancer and its treatment is needed to guide tertiary prevention of cancer.

Objectives: The SCALA (Survivorship after CAncer and its LAte effects) study will identify the major consequences of cancer and its treatment, their determinants, and their impact on the QOL of CS beyond the initial treatment phase.

Material and methods: SCALA will be embedded within two large nationwide French cohorts, Gazel and Constances, which include extensive baseline data (lifestyle, social factors, some QOL information), are updated annually, and are linked to France’s medico-administrative claims data (i.e., SNDS) for information on cancer diagnosis and treatment. Phase I of the SCALA study will analyse existing baseline data from the Constances cohort, and will compare QOL outcomes between prevalent CS and participants without a history of cancer. In Phase II, SCALA will holistically assess the QOL among CS included in both cohorts using newly collected data from the EORTC QLQ-SURV100, a questionnaire specifically developed to evaluate the full range of late and long-term physical, psychological and social effects of cancer in survivors of various cancer types and survivorship durations.

Perspectives: SCALA will provide a comprehensive mapping of cancer-related late-effects, their determinant and their impact on QOL. The findings will support healthcare providers in identifying CS at increased risk of late and long-term effects or impaired QOL and in tailoring long-term follow-up care to their specific needs. Moreover, the implementation of the EORTC QLQ-SURV100 within the population-based Constances cohort will constitute a major resource for the scientific community and for potential international comparisons.
Before start
SCALA PI: Dr Valerie McCormack
Institution: International Agency for Research on Cancer

For queries on the protocol or collaborations, please email Dr McCormack [email protected] or Dr Boucheron: [email protected]
1. Background & Rationale
Cancer dominates the disease burden in high-income countries. Fortunately, owing to improvements in early detection and treatment, survival has improved substantially over the past three decades for many types of cancer, thus the population of cancer survivors (CS) is rapidly growing. A third of the French population will be diagnosed with cancer before their 75th birthday (1). In 2017, there were an estimated at 3.8 million CS in France (2) (~6% of the population), many of whom would live decades beyond their original diagnosis (3). Most cancer research so far has focused on improving the length of life of CS. However, their quality of life (QOL) (4) is an important aspect of cancer survivorship encompassing physical, psychological and social well-being and remains understudied. CS’ QOL is often altered compared to individuals with no history of cancer (5) owing to a complex mix of the psychosocial experience of cancer and the possibility of it recurring (6), as well as the physical effects of past or ongoing treatment (7). Knowledge of the profile and determinants of the QOL and late effects of cancer, including their range, when they occur and how they differ between cancer types, is a first step to inform both CS and healthcare providers what timepoints in the post-cancer life-course need attention to particular late effects and QOL domains. Previous investment into cancer survivorship studies in France include the VICAN surveys (8), CANTO (9), and Seintinelles (10), which focused on common cancers only and often did not include a comparison group (CG) of comparable individuals with no history of cancer.
2. The SCALA research project
The present study of “Survivorship after CAncer and its LAte effects” (SCALA) will identify the major consequences of cancer and its treatment amongst CS in two large nationwide cohorts in France (Constances and Gazel), their determinants and their impact on QOL, overall and by cancer type. SCALA will use for the first time in France the new EORTC QLQ-SURV100 questionnaire (11, 12) – already translated into French – which was specifically designed to holistically assess the full range of late effects (physical, psychological and social) in long-term disease-free CS of multiple cancer types, survivorship durations and cultures. SCALA will address late effects and QOL beyond the initial treatment phase. Two major underlying health determinants will be explored, notably ageing and social inequalities. Direct (i.e., age, cancer type, type of cancer treatment received, and lifestyle factors) and indirect determinants (socioeconomic status, survivorship duration) of the QOL domains found to be altered in CS will be examined. We will include a comparison group of individuals with no history of cancer (CG) to CS, to disentangle cancer- from ageing-related impacts on QOL. Social gradients in the occurrence of late/long-term effects and QOL will be a specific additional focus.
3. Study aims
In SCALA, we will aim to investigate late occurring and long-term effects of cancer and its treatment, their determinants and impacts on the QOL among CS of all cancer types, overall and by cancer type. Specific aims (Figure 1) are:
                                                                                           
Aim 1. Through a CS vs CG comparison, to examine how sexual life, sleep, depression and healthy lifestyles are impacted in CS, overall, by cancer and sociodemographic factors.
Aim 2. To investigate cancer treatment and lifestyle determinants of the late effects and QOL domains affected by cancer, identified in aim 1, to inform avenues to improve QOL among CS.

Aim 3. In 2026, to administer the most up-to-date cancer-specific EORTC QLQ-SURV100 in two existing cohorts, providing data for aims 4-6.

Aim 4. To identify the range of late effects of cancer and its treatment and how these differ by sociodemographic and cancer-related characteristics (including subset external validation of late effects) among CS.

Aim 5. To examine whether and the extent to which the late effects identified in Aim 4 influence the functioning of CS.

Aim 6. To investigate factors influencing the occurrence of the five late effects with the largest burden amongst CS, as a means to inform prevention of late effects in CS.

Figure 1. Synopsis of SCALA aims and their inter-relationships

4. Methods
SCALA will be embedded in two existing large French cohorts (Constances and Gazel), which already have existing data for aims 1 and 2, annual lifestyle data, and, through access to public linked database, on cancer treatment information, whilst aims 4-6 will use newly collected QOL data (aim 3).
Study population

We will examine the late effects of cancer, their determinants and impact on QOL amongst CS at two time points (i.e., baseline and 2026) using either or both of two nationwide prospective cohorts in France (Constances and Gazel). While Gazel is smaller, it is older (13, 14) thus allowing investigation of cancer late effects in the elderly (~20,000 cohort recruited in 1989); however, Constances (15, 16) is larger (~207,000 recruited 2012-2021) and included a large population of prevalent CS (i.e. they had had a cancer diagnosis prior to inclusion).
Aims 1-2

In Aims 1-2, we will use the existing baseline data of the Constances cohort only, comprising ~10,000 prevalent CS – Constances eligibility criteria did not exclude participants based on health status – and 197,000 participants with no history of cancer (i.e., comparison group (CG)) participants; however, in Gazel, these were too few to be included in aims 1-2.
Identifying CS: All invasive cancers will be included in the group of CS. Cancer will be ascertained using a combination of self-reports of health problems over the last 12 months in the annual questionnaires, follow-up medical examination every 4 years (personal medical history reported by the physician in the baseline medical questionnaire only), and linkages to medico-administrative databases (i.e., cancer registries from occupational health services from 1989 to 2006 in Gazel only, and SNDS in both Constances and Gazel). The date of cancer diagnosis (i.e., month and year) – and thus age at diagnosis – and cancer type (ICD-10 codes) will be retrieved from the SNDS. In case a CS is not matched to the SNDS, the date of cancer diagnosis will be considered as the 1st July of the year in which the diagnosis was self-reported.  Precise information on stage at diagnosis is not available for either cohort; however, the SNDS will allow differentiation between metastatic and non-metastatic cancers. Cohort members diagnosed with in-situ cancers and non-melanoma skin cancers (NMSC, ICD10 C00-97, excluding C44) will be excluded from the group of CS as the treatment and psychological late effects are relatively minor in comparison to invasive cancers. Further exclusions (likely < 100 participants) will be those diagnosed with cancer in the two years before baseline, i.e. to exclude the acute survivorship period when treatment toxicities are more intense.

Age- sex-matched comparison group: All Constances participants who did not have a diagnosis of cancer (invasive, NMSC or in-situ) at the time of cohort baseline questionnaire completion will be included.
Aims 3-6

In Aim 3-6, we will invest in new data collection on QOL tailored to CS using the most up-to-date EORTC QOL questionnaire (i.e., EORTC QLQ-SURV100) administered to CS in both Gazel and Constances who are alive at the time of collection of EORTC QLQ-SURV100 data (expected in 2027). Despite the smaller contribution of Gazel than Constances to the total sample size, Gazel will be included as they will provide valuable information regarding late effects of cancer in elderly CS, which are often overlooked and poorly described, yet they form a majority of CS.

CS alive at new collection of QOL data: All respondents to the EORTC QLQ-SURV100 administered in aim 3 will be included. The other inclusion and exclusion criteria are similar to those for the prevalent CS mentioned above (please see section 4.2).
Existing data used for all aims

Demographic characteristics of volunteers - For both CS and CG cohort members, sociodemographic variables of interest for this study will include sex, marital status, life as a couple and menopausal status (only) to inform sexual life, and age (calculated based on date of birth in month and year). In CS, the latter will be used to derive age at (i) cancer diagnosis, (ii) completion of initial treatment, and (iii) completion of the EORTC QLQ-SURV100 questionnaire (aims 3-6). For women, menopausal status will be ascertained from baseline data for aims 1-2 and from follow-up data for aims 4-6, to the question “Do you currently use a means of contraception?” for which a possible answer is “I am menopausal”.  Furthermore, help with completing questionnaires will be used to identify volunteers with limited capacity to understand or fill in the questionnaires for any reason.

Cancer diagnostic data – see 4.2

Cancer treatment data - For each CS in both cohorts, the history of cancer treatment received will be retrieved from the SNDS. Variables considered will include type of treatment received (surgery, chemotherapy, radiotherapy, endocrine therapy, targeted therapy and immunotherapy), with dates of administration (month and year) from primary cancer diagnosis onwards retrieved from both the reimbursed ambulatory care (DCIR/DCIRS) and hospital care (PMSI). The pathology mapping of the SNDS (top cancer) will be useful to differentiate CS under active treatment from those who have completed it. As treatment history is complete from 2007 and 2011 onwards in Constances and Gazel, respectively, the treatment influences on QOL will be restricted to CS diagnosed from time onwards. For breast cancer, the type of surgery performed will also be considered (conservative or not, reconstructive).

Social and lifestyle determinants of cancer survivorship outcomes - For both CS and CG cohort members, potential non-modifiable determinants of cancer survivorship outcomes considered will be socioeconomic position at cancer diagnosis, sex,  age at baseline and at cancer diagnosis and for EORTC QLQ-SURV100 (aims 4-6), time since cancer diagnosis, since completion of initial and last treatment. For aim 6, as socioeconomic position (SEP) might have changed over time, and all Gazel participants will be retired at new data collection, we will use the closest available measure collected before cancer diagnosis. These SEP indicators will include:
  1. Occupation: Self-reported current employment situation and socio-professional category of the participant.
  2. Educational level - self-reported highest diploma obtained.
  3. Financial situation will be retrieved from self-reported personal and household income, and indicators of precarity (i.e., self-reported financial difficulties and/or foregoing care, self-assessment of current and future professional and financial situation, compulsory and complementary health coverage collected from the medical questionnaire and administrative information on Social Security affiliation retrieved from the SNDS, and EPICES scores retrieved from the medical questionnaire). These data will be useful to evaluate potential financial toxicity experienced by CS.
  4. As an ecological measure based upon place of residence (Constances), the Deprivation Index (Fdep) will also be used to capture broader socio-economic environment.

Modifiable lifestyle factors will be investigated in CS to identify actionable determinants of cancer survivorship outcomes. Lifestyle factors considered will include physical activity, body mass index (BMI, measured at baseline and self-reported thereafter), and consumption of alcohol, tobacco (including e-cigarettes) and cannabis (all of which are self-reported annually). For aim 6, to identify factors which might reduce the risk of late effects occurrence, we will use, for each lifestyle factor, the closest measure available from two years post-diagnosis onwards as representative of the average behaviour in the cancer survivorship period. To check this assumption, we will compare measures of lifestyle behaviours collected at different time points when these are available. Physical activity will be evaluated using calculated variables (physical activity off work, and physical effort at work) and self-reported sedentary and duration by intensity of activity self-reported at baseline and in annual questionnaires.
Existing additional baseline data required for aims 1-2

As the baseline questionnaire in Constances did not include a standard QOL questionnaire, we will conduct aims 1-2 using the following specific items and one depression scale self-reported in the health or medical questionnaires therein:
  1. Sleep quantity, quality, and disturbances;
  2. Physical health: comorbidities (personal medical history), self-rated perceived health and physical limitations;
  3. Depression from the 20-question Centre for Epidemiology Studies Depression scale (CES-D Scale);
  4. Sexual life: frequency of intercourse and satisfaction with one’s sexual life.
New data collection of QOL data (aim 3 for aims 4-6)

In aim 3 we will send the EORTC QLQ-SURV100 to all CS as a long-term investment for Constances, INSERM and the cancer survivorship research community. All eligible CS alive at the time of new collection of QOL data (expected in 2026) will be asked to complete the questionnaire. This questionnaire was specifically developed by the EORTC Quality of Life Group to capture the full range of late-occurring/long-term effects of cancer and its treatment and their impact on QOL of CS (11). It contains 100 items (see Appendix A) covering 12 functional and 9 symptom scales, a symptom checklist, 4 standalone items, and 10 conditional items; 73/100 items are included in the essential scales. The EORTC QLQ-SURV100 has been fully pre-tested, augmented and validated in a large-scale international field testing (current and last phase before its release) (17).

Figure 2. Structure of the EORTC QLQ-SURV100 questionnaire (adapted from Van Leeuwan et al (11)). N.B: the questionnaire comprises 9 symptom scales (S#), 12 functional scales (F#), 3 standalone items (X#), and 10 conditional items.

Logistics for the new QOL assessment: All eligible CS (cf. participant inclusion) will be sent the EORTC QLQ-SURV100 as part of their annual follow-up (expected 2026), with the option for online or paper completion of a questionnaire sent by post.
Online version: CS with an online account on the Constances platform will receive a letter by post followed by a personalised email containing a link to the online platform inviting them to complete a questionnaire. After secure log-in, CS will be asked to read the study information sheet before providing informed consent by ticking a box and proceeding to online questionnaire completion. To minimise missing data, answers to questions will be programmed to be mandatory for the online version with a “I do not want to answer” option. Upon completion, the data collected will first be uploaded on the platform’s cloud, and then securely transferred to the IARC research team. The data management team of the Constances platform will oversee the programming and sending out of the online version of the questionnaire.
Paper version: A paper version of the questionnaire will be sent to Constances CS who are not using that cohort’s online platform and to all Gazel CS. Personal details of the participants (names, addresses, cohort study identification numbers) will be provided to a trusted third-party (“Tiers de confiance”) as agreed upon by both IARC and Constances. Each eligible participant will be sent a personalised invitation letter with their cohort study identification number signed by both the cohorts and study PIs which will contain the information sheet, the consent form, and a pre-printed stamped envelope (T envelope) to ensure the completed questionnaire is safely returned free of charge for the participant. For Constances participants, a QRcode will also be included on the questionnaire in case they opt for online entry. We will ensure confidentiality of the status of CS (i.e., they might not have disclosed their personal history of cancer) is not broken (i.e. that the questionnaire does not identify the participant as a cancer survivor. This step will be discussed carefully between IARC, Constances and the study advisors. Dummy questions may have to be added at the start to give the option to answer having never had cancer, or wishing not to answer, even if the participant is being written to as a CS.

Reminders:  Reminder letters will be sent by paper post and email to all non-respondents, at 4 weeks (letter only) and 8 weeks (full paper-version study documents) after the first invitation.

Questionnaire design, printing and dispatch: All study documents will use the same visual appearance as the cohorts’ questionnaires. The information sheet will also contain information and contacts of cancer associations and psychologists, and social workers, in case a participant needs support. The printing and sending out of the questionnaire will be overseen by an external company compliant with the GDPR. Data entry of the returned paper questionnaires will be made by scanning and automatically reading the document. Concordance between the automatically entered data and the paper questionnaires will be checked on 20 randomly selected questionnaires.
QOL data (aims 4-6)

Late-occurring/long-term effects of cancer and its treatment: We will evaluate the 9 symptom scales of the EORTC QLQ-SURV100, the symptom checklist of 15 chronic effects of treatment – including musculoskeletal, metabolic, respiratory, vascular, skin, and neuropathic problems – fatigue, sleep disorders, bodily pain, health distress, negative health outlook, sexual problems, social interference and social isolation), and the three standalone items (financial difficulties, fertility, spiritual meaning of cancer). Each item of the EORTC QLQ-SURV100 will be assessed using a 4-point Likert Scale (“Not at all”, “A little bit”, “Quite a lot”, “Very much”). Items rated as “A little bit” or more will be considered as indicative of a late/long-term effect (12). Further outcomes provided in the annual self-reported questionnaires will be examined, which will include:
  1. Self-reported incident health outcomes first occurring two years or more after cancer diagnosis (i.e., post-treatment) to complement those reported in the EORTC QLQ-SURV100, such as: cardiovascular diseases (hypertension, non-fatal myocardial infarction and strokes), cardiometabolic diseases (diabetes, metabolic syndrome), respiratory disease, musculoskeletal, skin and digestive and urinary tract problems.
  2. Self-reported satisfaction scales regarding housing, neighbourhood, relationships, leisure, life, work, to obtain a better understanding of the CS with various areas of their lives.
  3. Self-reported life events which occurred within 12 months of the collection of the EORTC QLQ-SURV100 questionnaire, as these might have impacted the cancer survivorship outcomes reported.

Impact on CS’ functioning: We will evaluate the 12 functional scales of the EORTC QLQ-SURV100 i.e., physical, sexual, cognitive, emotional, role, and positive social functioning which represents the physical, psychological and social dimensions of QOL, as well as work, body image, positive affect, positive impact of cancer on behaviour towards others, positive health behaviour change and symptom awareness (Please see Fig. 2).
5. Statistical Analysis
Aims 1 & 2: In aim 1, we will perform a cross-sectional analysis of Constances baseline participants to identify which of the following four QOL/late effects outcomes (depression, sexual life, physical health and sleep) differ between CS (overall and by cancer type, time since cancer diagnosis and age at first cancer diagnosis) and their CG counterparts, whilst in aim 2 we will examine their determinants in CS only. For each outcome, a strong age-related pattern of missingness might be seen for sensitive questions (e.g., sexual life) thus the pattern of missingness will first be examined and where not completely at random, a multiple imputation method will be used to impute. Thereafter, regression models will be fitted using both the complete-case and imputation datasets. A further critical consideration for the CS–CG comparison will be to ensure adequate controlling for age. To do this, we will first restrict analyses to the overlapping age range where there are at least 20 participants in each 5-year age band in each of the CS and CG groups. Generalized linear regression models will then be fitted to estimate mean differences in each outcome between the CS and CG groups. Two approaches will then be used to control for age and their results will be compared. The first of these will be by adjustment and the second by constructing age-sex matched sets of CS-CGs. These approaches will first be used in aim 1 where CS-CG contrasts are examined for the four outcomes and by all strata of interest previously mentioned (cancer type, time since diagnosis, age, etc) including interaction tests.

Aim 2: We will use the same modelling framework but this time for a CS-only analysis, sex-age adjustment will be performed and the impact of treatment history and lifestyle factors will be analysed using a sequential adjustment approach to see whether they account for social gradients. Throughout, we will adjust for stage (metastatic or not) or a proxy based on ongoing treatment.

Aim 3: Prior to analysing the EORTC QLQ-SURV100 (Aims 4-6), we will first perform a summary of the cancer survivorship profile at the time of questionnaire completion, including the profile of CS responders and non-responders and predictors of response to assess for selection bias and representativeness. We will calculate the response proportion (responders/approached). We will also summarize the overall, sex-and age (at response)- and socioeconomic-specific distribution of CS in terms of: (i) the type(s) of cancer survived by the CS, ii) time since first cancer diagnosis, iii) time since last completion of treatment (excluding maintenance treatments), and (iv) the number of prior primary cancers.

Aims 4 and 5: In aims 4-6, we will perform a cross-sectional analysis of all CS respondents to the EORTC QLQ-SURV100 questionnaire. The score for each of the 9 symptom scales and of the 12 functional scales of the EORTC QLQ-SURV100 will be calculated as per the user manual, and will be treated as discrete variables (score from 0 (lowest QOL) to 100 (highest QOL) after standardisation). Each of the 15 symptoms checklist, rated from 1 (“Not at all”) to 4 (“Very much”) will be considered both as (i) binary – with a given symptom considered present if rated as “A little bit” or more by the CS – and (ii) categorical to examine its severity. We will analyse history of each self-reported health outcome considered in aim 4 pertaining to the existing data in Constances as a binary variable (ever/never). In aim 4, each (i) symptom scale, (ii) item (i.e., symptom and standalone) and (iii) newly onset disease will be described overall, and stratified by (i) age category, (ii) sex, (iii) SEP, (iv) major types of cancers (all those with at least 100 CS; those with <100 CS will be regrouped under an “other” group), (v) survivorship duration since diagnosis (i.e., [2-5[ i.e., short-term CS, [5-10[ i.e., long-term CS, and 10+ years i.e., very long-term CS), and (vi) type of cancer treatment received at any time, excluding potential maintenance treatment which might be taken for over a decade, e.g., endocrine therapy in breast cancer. As Constances and Gazel substantially differ in their study population, analyses will be stratified by cohort. Relationships between symptoms scales will be explored using scatter plots and structural equation models. In aim 5, the abovementioned analysis will be repeated for the 12 functional scales. Then, we will evaluate the effect of each one of the late-occurring/long-term effects identified in aim 4 by adding them separately to appropriate regression models to evaluate their relationship, for the symptoms (i) checklist and (ii) scales, with functional scales scores belonging to the same structural group of the EORTC QLQ-SURV100, and, for newly onset diseases post-cancer treatment, with all functional scales – considering no structural group membership assumption. Symptoms (i) checklist and (ii) scales and other self-reported health outcomes (from the existing data) associated with the functional scale of interest will be added to similar multivariate models to evaluate their independent impact on CS’ functioning of a CS. 

Aim 6: We will identify non-modifiable factors associated with a higher risk of developing one of the five late-occurring/long-term effects with the largest burden (aim 6a), and identify the extent to which modifiable lifestyle behaviours can reduce their burden (aim 6b).

Ascertainment of the five late-occurring/long-term effects with the largest burden: Of those identified in aim 4, the five cancer survivorship outcomes with the largest burden – i.e., those combining the highest prevalence/severity with the highest impact on CS’ functioning to generate a CS burden metric akin to a population attributable fraction in aetiology – will be determined using the following process: depending on their nature (discrete or binary) (i) the mean score of the symptom scale or (ii) the prevalence of the binary outcome (present/absent) (i.e., derived from a standalone item/symptom checklist or the self-reported health outcomes) will be plotted against the mean score of the functional scale impacted (or their combined average if multiple functional scales are impacted) (Fig. 3).

Figure 3. Identification of the five late effects for which needs are greatest in CS

The five cancer survivorship outcomes with the largest burden will be described according to SEP at cancer diagnosis (descriptions according to each of the other non-modifiable factors will have been performed in aim 4), and each of the four lifestyle indicators.

In aim 6a, univariate and multivariate regression models will be used to identify non-modifiable factors associated with each of the five outcomes of interest. In aim 6b, similar analysis will be performed to identify modifiable lifestyle determinants of the five outcomes of interest, unadjusted, and then adjusted for (i) non-modifiable determinants identified in aim 6a and (ii) all potential determinants (both non-modifiable and modifiable). Marginal methods will be used to predict the impact of modifiable determinants on both the reduction in (i) risk and (ii) severity of each of the five cancer survivorship outcomes of interest.
6. Ethical and regulatory aspects
Ethical approvals will be obtained from the IARC’s Ethics Committee (IEC), the French national ethics committee (CPP, i.e., comité de protection des personnes) and from the French national commission of information and liberties (CNIL, i.e., Commission nationale de l’informatique et des libertés). For aims 3-6, all CS will receive an information sheet describing the purpose of the Constance side study together with their right to withdraw at any time. Informed consent will be obtained. All results will be presented as aggregated data, thus not allowing to identify any individual.

Data will be handled as per Constances’ recommendations for data protection. The routing company in charge of printing and sending out the study material (aim 3) will be compliant with the GDPR. Constances, Gazel and IARC will solely have access to de-identified pseudonymized data. As an international organization, IARC is not subject to EU or any national data protection regulations but follows their own Data Protection Policy which follows United Nations regulations (18). For the present SCALA study, an access to the data is requested for 8 years.
7. Timeline & funding
SCALA has received funding from the INCA (grant no. INCA_19680) for a period of 4 years, which planned timeline is presented below in Figure 4.

Figure 4. Planned work for SCALA

Protocol references
1. Bray F, Laversanne M, Sung H, Ferlay J, Siegel RL, Soerjomataram I, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2024;74(3):229-63.
2. Institut National du Cancer. Panorama des cancers en France2024.
3. Botta L, Dal Maso L, Guzzinati S, Panato C, Gatta G, Trama A, et al. Changes in life expectancy for cancer patients over time since diagnosis. J Adv Res. 2019;20:153-9.
4. World Health Organization W. WHOQOL [Available from: https://www.who.int/tools/whoqol.
5. Paunescu AC, Preau M, Delpierre C, Jacob G, Pannard M, Delrieu L, et al. Quality of life among French breast cancer survivors in comparison with cancer-free women: the Seintinelles study. BMC Womens Health. 2024;24(1):17.
6. Magnani C, Smith AB, Rey D, Sarradon-Eck A, Preau M, Bendiane MK, et al. Fear of cancer recurrence in young women 5 years after diagnosis with a good-prognosis cancer: the VICAN-5 national survey. J Cancer Surviv. 2023;17(5):1359-70.
7. Dugué J, Humbert M, Bendiane MK, Bouhnik AD, Babin E, Licaj I. Head and neck cancer survivors' pain in France: the VICAN study. J Cancer Surviv. 2022;16(1):119-31.
8. Institut National du Cancer. La Vie Cinq Ans Apres un Diagnostic de Cancer. 2018.
9. Sandoval JL, Franzoi MA, di Meglio A, Ferreira AR, Viansone A, Andre F, et al. Magnitude and Temporal Variations of Socioeconomic Inequalities in the Quality of Life After Early Breast Cancer: Results From the Multicentric French CANTO Cohort. J Clin Oncol. 2024;42(24):2908-17.
10. Seintinelles. Seintinelles 2025 [Available from: https://www.seintinelles.com/english-version/.
11. van Leeuwen M, Husson O, Alberti P, Arraras JI, Chinot OL, Costantini A, et al. Understanding the quality of life (QOL) issues in survivors of cancer: Towards the development of an EORTC QOL cancer survivorship questionnaire. Health and Quality of Life Outcomes. 2018;16(1).
12. van Leeuwen M, Kieffer JM, Young TE, Annunziata MA, Arndt V, Arraras JI, et al. Phase III study of the European Organisation for Research and Treatment of Cancer Quality of Life cancer survivorship questionnaire. Journal of cancer survivorship : research and practice. 2023;17(4):1111-30.
13. Goldberg M, Leclerc A, Bonenfant S, Chastang JF, Schmaus A, Kaniewski N, et al. Cohort profile: the GAZEL Cohort Study. Int J Epidemiol. 2007;36(1):32-9.
14. Goldberg M, Leclerc A, Zins M. Cohort Profile Update: The GAZEL Cohort Study. Int J Epidemiol. 2015;44(1):77-g.
15. Goldberg M, Carton M, Descatha A, Leclerc A, Roquelaure Y, Santin G, et al. CONSTANCES: a general prospective population-based cohort for occupational and environmental epidemiology: cohort profile. Occup Environ Med. 2017;74(1):66-71.
16. Zins M, Goldberg M, team C. The French CONSTANCES population-based cohort: design, inclusion and follow-up. Eur J Epidemiol. 2015;30(12):1317-28.
17. EORTC. EORTC Quality of Life 2024 [Available from: https://qol.eortc.org/questionnaire/surv100/.
18. International Agency for Research on Cancer. IARC Data Protection Policy. 2021.