Introduction

Next generation sequencing offers new diagnostic opportunities in the context of rare diseases [1]. It leads to a primary finding (PF) more frequently than other conventional strategies, but it can also reveal variants unrelated to the patient’s disease. These findings are either “incidental” when they are discovered by chance (IF), or “secondary” if they are actively sought (SF). Their discovery can lead to preventive or therapeutic action as part of a personalized medicine approach. The aim is to prevent the onset of disease, and to reduce morbidity and mortality. In 2013, the American College of Medical Genetics and Genomics (ACMG) defined a list of 56 genes accessible for prevention and/or treatment, including genes predisposing to cancer, cardiovascular and metabolic diseases. This list is regularly revised, most recently to 81 genes in 2023 [2,3,4,5,6]. Generation and subsequently accessibility of secondary findings (SF) remains a matter of debate in France and elsewhere. The European Society of Human Genetic and Canadian College of Medical Genetic have recommended that IF/SF not be made available until evidence of medical utility is demonstrated [7,8,9,10].

The international literature, in particular from the US, includes several quantitative and qualitative studies of patients, mostly in hypothetical situations, but less often in a context where the IF/SF are actually reported to patients [11,12,13,14]. All of these studies indicate that there is a broad desire to be informed of such results, whatever their category, to make life choices guided by this medical information. Conversely, professionals stress the risk of psychological consequences and the importance of establishing national recommendations, particularly for the definition of the “actionability” of genes [15,16,17].

Here we present our mixed study reporting the expectations, needs, and mid-term outcomes in patients/parents who chose to access SF after undergoing ES.

Materials and methods

General study design

The FIND study, conceived by a multidisciplinary team (geneticists, genetic counselors, psychologists, health economists, ethicists, sociologists, anthropologists, epidemiologists, patients’ associations), uses a mixed design associating qualitative and quantitative methods. The FIND study received approval from the local ethics committee (N°: 17-DIJO-01, 11/10/2017), and all participants provided consent for the use of research data in an anonymous manner.

List of studied secondary findings

In this project, SF was defined, according to ACMG and the French Agence de la Biomédecine (ABM) [18] recommendations, as pathogenic or probably pathogenic genomic variants (classes 5 and 4), that they have no direct relationship with the initial indication or, for phamacogenetic, variant-drug combinations with strong scientific evidence, that are actively sought by analyzing a pre-established list of genes. The methodology for interpreting SF has already been published [19]. Three actionable groups were considered (Supplementary Table 1): (i) 1st SF category 1 (SF1): variants involved in a genetic predisposition to a preventable or treatable disease, mostly occurring in adulthood, using the list of 118 genes proposed by Dorschner [20], including the ACMG list of 59 genes [2, 3], plus four genes involved in bleeding disorders; (ii) 2nd SF category (SF2): variants in 114 genes with an impact on genetic counseling for future pregnancies, including heterozygous carriers of a recessive disease whose prevalence of heterozygotes would lead to screening of the partner in a given population and female carriers of an X-linked recessive disease; (iii) 3rd SF category (SF3): two genes whose homozygous or compound heterozygous pharmacogenomic variants have a clinical interest for drug adaptation in the context of neurodevelopmental disorders.

Fig. 1: Expected study design of the FIND study.
figure 1

A mixed methodology was proposed, including both a quantitative and qualitative study. For the quantitative study, all participants had to fill in specific and standardized questionnaires at inclusion (QT0). In order to have 250 respondents, with a 25–30% risk of being lost to follow-up, the number of subjects to include was 330. They also had to fill in questionnaires at time of results (QTR), 6 months after results (QT6), and 12 months after results (QT12). QT0: study data concerning the perception of the interest of SF, the needs and expectations regarding the information given about SF, the selected SF category, as well as the reasons for these choices. QTR - QT6 – QT12: The self-administered questionnaire consisted of closed multiple-choice questions and visual analog scales (VAS) that ranged from 0 (lowest agreement or satisfaction score) to 100 (highest agreement or satisfaction score). Standardized scale: anxiety (STAI-Y A and B), depression (CES-D), and quality of life (QoL; SF12 mental and physical). For the qualitative study, the same 30 participants with SF+ were asked to participate at TR, T6, T12, and 10 participants with negative results were planned to be interviewed at TR. Participants also filled in the questionnaires.

Full size image

Study course (Fig. 1)

Fetuses, children, and adults with developmental abnormalities requiring solo ES in order to identify the causal diagnosis were recruited from October 2017 to March 2019 in three French genetic clinics (Dijon, Lyon, and Paris). Participants surveyed were the parents and all included adult index cases able to understand, speak French and consent to the study. The geneticist informed the patient/the parents about the project and about the search for SF. If they were interested in participating in the study or having additional information regarding SF to make their decision, they systematically met with a genetic counselor specifically trained in genomics. At the end of this information process, within a 1-month period, participants confirmed their desire to access SF or not (SF1, 2 and/or 3) and therefore their inclusion in the study. If they did not wish to access SF results, participants were asked what led them to their decision. Participants could change their mind at any time until the results were reported. The inclusion could not be confirmed if one parent did not wish to access SF.

First assessment (inclusion): in the month following inclusion, all included participants were asked to complete standardized scales assessing anxiety (STAI-Y A and B) [21], depression (CES-D) [22] and quality of life (QoL ; SF12 mental and physical) [23], administered at each assessment, and self- administered questionnaires (QT0) (paper or online) to collect socio-demographic data, clinical data, and expectations regarding SF (based on multiple-answer questions).

Second assessment (ES and SF results delivery—time of report (TR)) When ES and SF results were made available, a consultation with the geneticist was scheduled for each patient. During the consultation, the geneticist first reported the PF and secondly, if participants confirmed their choice to access SF, the SF results were delivered by the geneticist and/or the genetic counselor, assisted or not by another specialized physician. Then participants completed the same three standardized scales of anxiety, depression and quality of life and a specific self-administered questionnaire (QTR) according to the nature of their result (with or without PF, with or without SF, and according to the SF category). These questionnaires, based on multiple-answer questions, explored the reasons behind their choice to access SF, possible changes of opinion, their experience and satisfaction with the results obtained and the information and support received.

Thirty parents of minors or of protected adults with a SF (SF+) (all SF group 1 and a selection of SF groups 2/3) and ten without SF (SF-) were asked to participate in a semi-directive interview with one of two training psychologists or one training sociologist according to center in order to evaluate their feelings after the results were reported.

It was decided that this research would focus primarily on the experiences of parents/patients who had received a SF1 (predisposition to late-onset actionable diseases), by offering an interview to all patients, given their limited number, and their major interest since they include the minimum list defined by the ACMG. Thus a selection of files were made for patients with SF2 (genetic counseling) or SF3 (pharmacogenomics), taking care to explore a variety of situations: Patients with a positive or negative primary diagnosis; SF diversity (e.g., CFTR and HFE for SF2); minor and adult patients; mother or father interviewed.

The interview guide was developed jointly by a sociologist and a psychologist following their SEQUAPRE experience [11]. Most of the first interviews took place in person following the consultation dedicated to the delivery of results of ES.

Follow-up assessments (6 and 12 months after delivery of results—T6 and T12) Each individual with SF+ was asked to complete specific self-administered questionnaires in addition to the same standardized scales of anxiety, depression and quality of life, at 6 and 12 months after the delivery of results (QT6 and QT12). These questionnaires explored the experience at a distance from the results, the perception of the interest of the results, the satisfaction, the needs and expectations in terms of support, the level of appropriation of the results, and the consequences of the results for the patient and his family in terms of prevention and care, basing on multiple-answer questions. The same parent of individuals with SF+ was interviewed at each timepoint in order to explore how the impact of SF results on the individual and the family changed over time.

Data collection and treatment

All the questionnaires were anonymized and entered in a specific database via a Clinical Data Management System (Cleanweb®, Telemedecine Technologies, Boulogne-Billancourt, France) managed by data managers from CIC-EC1432. The interviews were recorded, transcribed in full and anonymized, and then analyzed after several in-depth readings to identify emerging themes that were encoded and analyzed with the NVivo software (version 11). To reduce analysis bias, the first five interviews were analyzed and coded jointly by a sociologist and a psychologist trained in qualitative analysis.

A first conceptual thematic encoding was carried out based on the four concepts emerging from the literature review: psychological effects of the announcement, projection into the future, effect on parenthood, repercussions on lifestyle. Sub-themes then rapidly emerged and a final level of sub-themes was sometimes necessary.

Statistical analysis

Continuous variables were expressed as means ± standard deviations (SD) or medians and minimums-maximums according to their distribution, and categorical variables as frequencies and percentages. At each time, means were compared using Kruskal-Wallis analysis of variance test when appropriate. Normality of variable distribution was checked with Shapiro-Wilks tests. Proportions were compared using Chi-squared or Fisher’s exact test when required. Longitudinal analyses of depression, anxiety, and quality of life scores were performed using linear nested random-effect models.

A p < 0.05 was considered statistically significant. Analyses were performed using SAS ™v9.4 (SAS Institute, USA) and Stata v15.1 (StataCorp LLC, USA).

Results

Characteristics of the population included

Over the study period, 435 patients undergoing ES were given the option of adding a search for SF. Finally, 78% (340 patients) requested to access SF (Fig. 2 and Supplementary Table 2).

Fig. 2: Flow chart of patients and participants.
figure 2

A This figure shows the number of patients eligible for the study and the number of people who wished to access their SF results. For people with SF, the SF category is specified; B Number of completed questionnaires and interviews (results of questionnaires and interviews with parents without SF are presented in Supplementary Table 3).

Full size image

Included patients were an average age of 13 ± 11 years and 56% were males. They presented with intellectual developmental disorder (IDD) (71%) and/or malformations (72%), and the majority (73%) had undergone previous genetic investigations. The characteristics of participants who completed questionnaires are presented in Table 1. Sixteen adult patients including six with learning disabilities (all interrogated independently from their parents), 294 mothers and 180 fathers completed at least one questionnaire. The level of education was at least equal to a high school diploma in 61% of mothers and 59% of fathers. One third of the mothers were not working, 76% by choice. Among the parents who were a couple (72%), 25% were in a precarious socio-economic situation (Epices score above 30). Among single parents, 58% of mothers and 29% of fathers were in a precarious situation. The level of situational anxiety was higher among mothers (27%) than among fathers (15%) (p = 0.003).

Table 1 Characteristics of participants and results of standardized and specific questionnaires at inclusion (n = 490).
Full size table

Characteristics of the population interviewed

Thirty parents (21 mothers and 9 fathers) of 26 patients (21 minors and 5 protected adults) with SF+ took part in a semi-structured interview (at TR, 13 with SF1, 11 with SF2 and 5 with SF3; Fig. 2). Eighty-three interviews were conducted and analyzed (29 at TR, 28 at T6, 26 at T12; 59 with mothers, 24 with fathers). Only two attributes were notable: the nature of the SF rendered and the time of analysis (Fig. 3).

Fig. 3: Evolution of psychological aspects based on the analysis of interviews according to the type of SF and the time of results.
figure 3

A Worry and anxiety; B Feeling at risk for participants or their family members.

Full size image

SF results

Overall, SF were identified in 47 patients.

Among SF1, 16 SF+ were reported. Variations were found in genes conferring a genetic predisposition to cancer, cardiac, hematological or metabolic diseases. For SF2, 20 SF+ were reported, of which 75% concerned HFE gene. For SF3, 11 homozygous or compound heterozygous variations were reported (Table 2). The family segregation is not available for three patients.

Table 2 Result of SF analysis and specific questionnaires at TR, T6, and T12.
Full size table

First assessment: results of standardized scales and specific questionnaire before ES (QT0) (Table 1)

Most participants chose to access all three SF categories (94%; n = 323/340). Overall, 85% of participants had specific expectations regarding SF, mainly to implement preventive measures (77%) in order to prevent possible risks (73%), to better anticipate the future (77%), and to improve their health status (77%). Participants were ambivalent regarding the hypothetical reporting of SF for conditions with no treatment or prevention options and for SF with an uncertain status, with a tendency to wish for more results in the absence of actionability when the adults concerned are themselves questioned. However, given the small number of adults questioned, these results should be treated with great caution.

When asked how they had made their decision to access SF, 56% decided with their geneticist, 35% alone, and 17% with their partner. According the qualitative data, the choice was considered difficult for a majority of participants even though they expressed a strong motivation.

Second assessment (TR) and follow-up assessments (T6 and T12)

Medical actionability: a strong and unanimous expectation at time of report and during follow-up (Table 2)

At TR, the parents were globally satisfied by the offer to access SF results (94%), especially if a PF was found (100%). At T12, satisfaction remained high (92%). More specifically, at TR, parents whose child received a SF1 were satisfied that it would lead to better monitoring (86%) and allow them to inform their family (67%). Parents whose child received a SF2 were satisfied overall, but the reasons were diverse: to take an active role and to make responsible decisions with the available information (42%), but also to access to genetic counseling for future pregnancies (35%). The parents whose child received a SF3 were satisfied because their doctor could adapt treatments (81%), to avoid complications (56%), and because they could adopt an active role and make responsible decisions with the available information (50%).

For the interviewed parents, the medical actionability of the SF corresponded to a strong, unanimously expressed expectation that it would motivate their choice of access and confirm it after the report. Parents of children with disabilities, whose sense of responsibility is particularly strong, had a more pronounced desire to know whether their child had genetic risk factors in order to be able to better anticipate, prevent, act and protect their child.

Anxiety, depression, and quality of life

When considering all responding parents (whatever the PF and SF results), the analyses of the standardized scales showed that mothers tended to have significantly higher levels of anxiety and depression and lower quality of life compared to fathers, whatever the time of assessment (p < 10-3 for all score results). In addition, independently of the responder’s sex, having a PF was associated with increased anxiety and depression levels (+2.8 ± 1.2 points, p = 0.014; +3.2 ± 1.5 points, p = 0.036, respectively). Conversely, the fact that a SF was revealed was not associated with increased scores of anxiety and depression (pmin = 0.236). When examining specifically SF+ situations, the gap between anxiety and depression levels for mothers and fathers persisted over time (p = 0.037 and 0.015, respectively). Notably, the physical component summary (PCS) of the SF12 scale also appeared to be deteriorated in mothers, particularly when SF1 were assessed.

The interviews showed that SF1 were a source of worry and anguish to a greater extent and for a longer period of time than for the other SF groups (Fig. 3) (‘As soon as you say “hypertrophic cardiomyopathy”, everything goes on alert.’ Mother, SF1, T12, MYBPC3 variant detected in her daughter). Moreover, the announcement of a genetic predisposition to cancer or heart disease activated representations of death (“My mother’s older brother died one morning in his bed, just like that. So I asked myself the question: Could it have been hypertension that took him? We don’t know.”Mother, SF1, T6, SDHB variant in her and her daughter, leading to the discovery of hypertension in the mother).

The anxiety and anguish caused by the announcement of a SF2 was particularly apparent in some contexts (pregnancy in progress, upcoming surgery) or due to disease representations (“It’s horrible. It’s a slow end. It’s witnessing the agony of a child, it’s appalling.”Mother, SF2, TR, CFTR, variant detected in her daughter), especially at the time of the results announcement, although this effect faded over time. On the contrary, the reporting of SF3 was not anxiety-provoking: “It’s of no great consequence” (Mother, SF3, TR, CYP2C9).

Parental concerns and anxiety were maintained over time because the risk factor announced concerns the child first, but then potentially the parents themselves, and then other family members, especially siblings (“The anguish with regard to my other children, to tell myself that I hope I didn’t give it to them too.” Mother, SF1, TR, DSG2 gene). Having to disclose the findings to the rest of the family was also anxiety-provoking, depending on the situation.

Nine out of 30 parents recalled the shock caused by the SF announcement and in a more significant and prolonged way for SF1 (“It was announced, telling me that I was like my daughter, at high risk. I had flipped a coin. I didn’t end up on the right side of the coin. It was like a cold shower.” Mother, SF1, T6, BRCA2 variant carrier like her daughter). In retrospect, half of the parents said it would be impossible to anticipate the return of a positive SF result and its consequences.

Specific situations where the discovery of SF1 had a psychological impact have been summarized in Table 3. In three patients, psychological follow-up was implemented given the issues generated by the SF result, added to an already medically complex situation, and in a particularly precarious social situation. The SF result was not the central reason for this follow-up but it added another layer of vulnerability.

Table 3 Detailed presentation of the interviewees’ situations.
Full size table

Impact on lifestyle

In the questionnaires, the parents in SF1 revealed that there had been major changes in care and lifestyle for their child (38% at T6 to 46% at T12), including the implementation of preventive measures (e.g., sport or diet) (44% at T6 to 38% at T12).

Change in feelings

Although some interviewed parents said they felt serene, confident and reassured, others expressed ambivalent feelings at later points in time (for SF1 and SF2 in particular). These participants measured both the benefits of the search for SF and the associated risks (“I am quite satisfied with the fact that I have an answer, and at the same time, I am a little apprehensive about what can happen afterwards, about anesthesia, possible surgery…” Father, SF1, PROC gene, T12). The importance of medical and psychological support before and after the decision-making process also emerged in some parents’ later comments.

After the reporting of results, the carriers acquired a new status of: “being at risk” [24].They no longer felt fully healthy but they were not yet ill. This liminal stage was drawn out by uncertainty and doubts. We observed a growing and sustained state of confusion over time for individuals who had SF1 results more than for SF2 and SF3 (Fig. 3) (“There was something we didn’t know, and now we have this risk in the family. We live with it, it’s something we don’t see but we live with that risk in the family. Now we know it’s there. Before, we didn’t know that we had something like that in our body” mother, SF1, SDHB gene, T6).

Later on, for SF2 and SF3, some participants could not remember the results (7/29), indicating that they were not particularly concerned about SF that had a limited effect. For others (10/30), there was confusion with the PF, raising questions about the difficulty of understanding such complex results.

Finally, all of the parents who responded to the questionnaires and 28/30 interviewed parents said that they had no regrets about accessing SF results (’It still allows you to be able to protect and preserve. Make sure everybody’s okay. So no, I have absolutely no regrets. And tomorrow if I’m asked to do it again, I’ll do it 100%’ Mother, SF1, DSG2 gene, T12). Twelve parents out of 30 put forward the fact of being able to anticipate thanks to the SF (“For this Marfan disease I’m not more worried than that because there’s a very correct follow-up we’ll say, we’re well supervised, everything has been explained to us correctly.” Mother, SF1, FBN1 gene, T6; “It’s useful, afterwards we’re lucky, we’ve come across a disease that’s easy to treat, and what’s more easy to detect, so it’s interesting to have had the information all of a sudden.” Mother, SF2, HFE gene, T12”).

Discussion

Our study design was developed in an attempt to asses, using a forward-looking approach, the many issues raised by SF in a since only the return of IF to patients is currently authorized in routine care by bioethics laws in France, while SF is still prohibited outside the research context [25]. Access to SF is seen as a change of clinical paradigm because it offers patients an opportunity to access results beyond their initial request. Furthermore, this offer is contrary to the principles and protocols that have been implemented worldwide to best manage sensitive requests for presymptomatic diagnoses. Nevertheless, because this is a major issue of interest in the development of genomic medicine, research studies have particularly been encouraged to discuss the evolution of positions in the future. In some respects, the results gathered in this study of SF can also be transposed to IF.

For the FIND study, we chose a mixed design combining standardized anxiety, depression, and quality of life scales with specific questionnaires studying the psychosocial impact (satisfaction, worry, and changes in life habits) and qualitative interviews allowing us to refine these different aspects. We felt this approach was essential if we were to properly interpret the results in families that have been severely affected by a rare disease. The collection of this data again at 6 and 12 months allowed us to assess the mid-term effect of the reporting of results. We showed that, while the reporting of SF results had little influence on the standardized scales, but it could have real repercussions at the individual level, particularly when a predisposition to late-onset actionable diseases was revealed. Although we expected the SF1 to have a greater psychological impact than the SF2 and SF3, we found that occasionally there were psychological effects related to particular situations, particularly among women. This study confirms that quantitative data alone are not a good marker of the impact of the SF results seeing as they obtained some answers but without any reference to the subject’s personal history. Therefore, a qualitative assessment of the interviews was essential for identifying how individuals were affected by SF in a contextualized and understandable way. The qualitative study suggested, for example, that surprise and shock may be linked to an impossible anticipation, since they were not sufficiently psychologically prepared to get these results.

The discourse of a mother who has breast cancer but no genetic predisposition to breast cancer identified in her child (“not to have passed it on!” Mother, SF1, LDLR gene, T6) alerts us to the sometimes unrealistic expectations (knowing or eliminating all risks), with the risk of false reassurance of negative result. Indeed, a false-negative result could lead patients at risk of optimism bias [26], to false reassurance [27].

Some of our results are consistent with those of previously published real-life studies (3/6 from North America) evaluating the impact on patients in the short to mid-term (Table 4) [28,29,30,31,32,33]. Only one study had a mixed design; the others were quantitative or qualitative. In general, wanting to access SF was motivated by medical actionability [28]. Our study showed that the lifestyle repercussions and parents’ perception of their children was different in SF1 (actionable late-onset disease) and in the two other groups. Consequently, despite the psychological and social issues, most studies, including ours, found that no or few patients regretted accessing their SF results [17, 29,30,31,32,33,34].

Table 4 Review of other real-life studies.
Full size table

Despite positive feedback overall, the nature of the announced SF appeared to be a determining factor in the intensity of parental worries and anguish, especially depending on the nature of the representation caused by the result (representation of death associated with cancer and cardiovascular risks). Similar to our study, previous studies showed that there is not a severe psychosocial impact for participants [29,30,31], and that the individual’s previous state of anxiety, history and the type of SF seem to be the factors that most influence how they experience the announcement [28, 29, 33, 35, 36]. Interestingly, the crosscheck of quantitative and qualitative data confirmed that the initial psychological status was an important prognostic indicator for the participants’ reaction when the genetic results for SF were reported. Accordingly, parents with history of depression and/or a high-level of anxiety had a higher risk of psychopathological consequences following the announcement of a SF for their child [31].

Also, we believe that it will be essential to provide additional support for individuals in situations identified as at-risk. With the current resources, if SF were to be offered to all persons undergoing ES/GS, it would be impossible to provide specific clinical consultations and systematic psychological support in each case. However, at-risk patients should be identified and psychological support offered if the patient is in distress.

Otherwise, this study revealed some SF actionable in adulthood among minors, providing relevant information for the children’s parents/families. At the same time, due to the organization of French health care, and without a national shared medical record, we must anticipate that contacting the proband in adulthood could be challenging. In addition, when Feinberg [37] insisted on the importance of preserving the “right for a free future”, he underlined that it is necessary to ensure the quality of the early child-parent relationship while protecting the child from too much knowledge. Richer and Laberge proposed that the list of accessible SF could be revised to contain only diseases actionable from childhood [38]. We recognize that this was a limitation in the design of the FIND study, therefore our team is now conducting a new study, ancillary to the DEFIDIAG study (DEFIDIAG-DS), proposing trio GS in patients with undiagnosed ID [39]. In DEFIDIAG-DS, the search for SF (ACMG list version 2) is offered to parents but not to their child. Results will be expected in the future.

Finally, the search for SF can be similar to a pre-symptomatic testing process without information regarding family history. Studies on this subject can help us to establish a framework by better measuring the psychosocial consequences of reporting SF [35, 36, 40]. To extend our understanding of this subject, we have prolonged the follow-up for patients in whom SF was identified in the FIND study over a period of 10 years, which should allow us to better evaluate the long-term effects and to measure the consequences in terms of access to care and on the family as a whole.

In conclusion, using data obtained from standardized scales, questionnaires, and interviews, the FIND study revealed that the diagnostic reporting of a genetic SF result is not a trivial matter and that there are mid-term consequences on psychological health, care pathways, and lifestyle. Although participant satisfaction was high and, in most cases, serious incidents were not observed, care should be taken to avoid putting patients in situations of vulnerability and, as much as possible, patients need to be guided and supported when choosing to generate and access this type of presymptomatic screening, especially after the results are delivered. This study, thanks to its mixed methodology and the fact that it took into account the psychological state and anxiety of patients prior to the reporting of results, can be used as a point of reference for countries that have not yet decided on the question of offering access to SF in patients undergoing ES/GS.