Cancer Patients Who Experienced Diagnostic Genetic Testing for Cancer Susceptibility: Reactions and Behavior after the Disclosure of a Positive Test Result¹

Recent advances in molecular genetics have led to the identification of susceptibility genes for hereditary cancer syndromes. Breast ovarian and colon cancers (∼5–10%) are inherited (1). Two genes, BRCA1 and BRCA2, have been identified as susceptibility genes for HBOC³ (2). The DNA mismatch repair genes (MMR genes: hMLH1, hMSH2, hPMS1, hPMS2, and hMSH6) are responsible for HNPCC (3).

Families at risk for whom inherited cancer is suspected now have the opportunity to undergo genetic testing (4). Diagnostic genetic testing in a family should begin with the case index, defined as the subjects affected with cancer who took part in the initial molecular research. If a deleterious mutation is detected in known cancer-predisposing genes, predictive testing could then be performed to determine the genetic status of each unaffected family member and predict their risk of developing cancer (1).

This new procedure in oncology practice raises many questions about the potential psychological, social, and ethical repercussions and also the real benefit for the patients and their families (5, 6, 7). Thus, it is essential to evaluate the repercussions of genetic testing in families (4, 8, 9).

To date, most of the published studies have focused on the impact of predictive testing in healthy individuals (10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20). Only a few studies have concentrated specifically on the psychosocial repercussions in patients with cancer, and these are a priori studies focusing on interest in genetic testing or intention to learn results (21, 22, 23, 24).

However, the case index patients will be the first and only individuals to be informed about the existence of a mutation in their family. As a consequence, the patients will have to inform their relatives about the result and the availability of predictive testing, because diagnostic genetic testing does not concern a single individual but the whole family. In addition, those patients with a personal history of cancer who have been identified as carriers of a cancer-predisposing mutation have a high risk of developing a second cancer (2, 25). For these reasons, they may be vulnerable to distress.

We performed a prospective evaluation of the consequences of a positive result disclosure in case index patients identified as carriers of a breast ovarian- or colon cancer-predisposing mutation. The aim of this study was to gain insight into the experiences of cancer patients after diagnostic genetic testing through the psychological impact of a positive result and their role as a messenger in their families; therefore, we can define optimal strategies for care and guidelines.

Eligible subjects for the study had a personal history of cancer and had been identified as carriers of a susceptibility gene for breast or colorectal cancer through two cancer-predisposing testing programs currently underway at the Center Léon Bérard, a regional comprehensive cancer center: one offering BRCA1 and BRCA2 testing in HBOC syndrome and one offering hMLH1 and hMSH2 testing in HNPCC syndrome (26, 27, 28).

The procedure for genetic counseling, which was the same for these two testing programs, is outlined in Fig. 1. When an HBOC or HNPCC syndrome was suspected after the evaluation of the pedigree by the geneticist, genetic testing was offered to adult subjects affected with cancer in the family. All participants provided written informed consent before a blood sample was taken. When a mutation in a cancer susceptibility gene was identified, the patients received a letter inviting them to an individual result disclosure session. This session was animated by a geneticist using a standard procedure (following written guidelines) to ensure that participants received uniform information, which included explanations of their genetic status and the implications of this in personal and familial terms and a review of the medical recommendations. At the end of the session, they were invited to participate in this study except when the patient had advanced disease with a life expectancy of <3 months, for ethical reasons.

From June 1995 to December 1999, 104 HBOC families and 50 HNPCC families were fully analyzed through the genetic testing programs ongoing in our institution. A clearly deleterious mutation was detected in 25 subjects from 17 HBOC families and 31 subjects from 20 HNPCC families. The results have been disclosed to only 31 of these 56 positive subjects because 13 of them had died by the time the result became available, and the remaining 12 subjects did not respond to our invitation. From June 1998 to June 2000, 23 of these subjects were included in the present study because 4 individuals with rapidly evolving disease were not eligible, and 4 others refused to participate. The characteristics of all of the tested subjects are summarized in Table 1.

The consequences of the result disclosure were assessed 1 month after receiving the positive test result by a 1-h semistructured interview. All of the interviews were conducted by the same clinical research nurse from the Center Léon Bérard psycho-oncology unit, which was not a member of the genetics team. Immediately after the interview, the patients were asked to complete the HADS self-questionnaire (29) to detect high levels of anxiety and/or depression, likely to require appropriate psychological management. After the interview, a second appointment with the geneticist was performed to complete the genetic counseling (Fig. 1).

The interview schedule included the following four major topics: (a) the personal feelings and reactions of the patients before and after the disclosure of the result; (b) the meaning they give to the result for themselves and their family; (c) their opinion about the pros and cons of genetic testing; and (d) the communication of information about genetic testing and its effects on other family members. These themes were identified from a literature review on genetic testing in cancer and through discussion with the relevant clinicians or researchers during the working sessions of the oncogenetic multidisciplinary team in Lyon (i.e., all of the authors except V. B.).

For each topic, the semistructured questionnaire consisted of one to five open questions to elicit spontaneous responses from the patients (see appendix). The exact words used by the patients were noted at the time of interview, transcribed afterward, and further analyzed by one of the authors (V. B.). The responses were grouped into similar items and then classified according to their implicit positive or negative impact.

Each of the open questions was followed by closed preset subquestions with predetermined responses to obtain more precise responses or complete the information if needed.

The HADS includes 14 questions, of which half establishes an anxiety score, and the other half establishes a depression score. It has been suggested that a combined score (the sum of the scores from the two subscales) of >13 indicates an adjustment disorder (30). An anxiety score of >10 indicates abnormal anxiety, and a depression score of >8 indicates abnormal depression (31).

The interview was performed on average 6 weeks (range: 4–13) after disclosure of the result. The median duration of the interviews was 70 min (range: 45–80). The interviews went well for all but 1 patient who became irritated because there were too many questions.

In reporting the data, the percentages for some of the responses are provided with a summary of the data, despite the small sample.

Table 1 summarizes the socio-demographic and clinical characteristics of the 56 subjects who were tested from the 37 families with a mutation. The 23 patients included in the study were mainly women (74%), and their mean age was 47 years (range: 27–72). All patients except 1 (a woman aged 27) had had children. The mean age of the children at the time of the evaluation was 23 years. Twelve patients (52%) had at least one child >18 years old, thus potentially concerned by predictive tests. Thirteen patients (57%) still had at least one parent alive. The mean number of cancers in the close relatives (first and second degree) was five per family (range: 1–12).

Six of the 7 women who were identified as carriers of a BRCA1 mutation had developed breast cancer, and 1 developed an ovarian cancer. Among the other 16 patients who carried a mutation of a MMR gene, 8 women and 5 men had developed colorectal cancer, 2 women had developed endometrial cancer, and a man was diagnosed with urinary tract cancer. The mean age at diagnosis of first cancer was 40 years. Seven patients had developed a second primitive tumor. All patients have undergone surgery, and most of them have received chemotherapy (61%). At time of study, all but 1 of the patients had nondetectable disease and were not undergoing treatment. The median time since diagnosis of last cancer was 48 months.

Patients who did not respond to our invitation for disclosure of the results were comparable with the included patients for age at diagnosis of first cancer, sex, type of hereditary syndrome, history of first cancer, number of children, and status of the disease at time of study. However, they were more likely to be older at time of study (median age of 63 versus 47 years, P = 0.006), have a longer time since last diagnosis of cancer (156 versus 48 months, P = 0.006), and were more often diagnosed with second primary tumor (58 versus 30%, P = 0.1) and with locally advanced tumor (42 versus 9%, P = 0.05) but had less often received chemotherapy (25 versus 61%, P = 0.04).

As expected, patients who were considered to be ineligible for the study or had died by the time the result became available were more recently diagnosed with cancer and more often with a second primary tumor and/or metastatic disease and treated with chemotherapy. All these patients except 1 were women, and most of them presented an HBOC syndrome.

During the week after the disclosure of their test results, 12 patients (52%) spontaneously expressed no major emotional changes. They said they thought they were carriers of a cancer-predisposing gene before result disclosure. Conversely, 8 patients (35%; 3 missing data) expressed distressed reactions, stating: “You no longer feel cured,” “The anguish of cancer comes back,” “I was sad and unhappy,” and “It was the downward slope.” The details of the different emotional reactions that patients felt are shown in Table 2. The majority of the patients expressed no major negative emotion. Twenty patients (87%) did not express more than one negative emotion, 21 patients reported that they were not surprised, and 20 patients reported that they were not shocked. Twelve patients said they were reassured. Dissatisfaction and discouragement were reported each by only 5 patients. However, 7 patients reported that they were unhappy, and 10 said that they felt worried. Overall, 14 patients (61%) expressed at least one of these last four negative reactions. Among these patients, 8 patients said they had never (3 patients) or rarely (5 patients) thought about genetic testing during the time of molecular research and were not worried about genetic testing at that moment. They stated that they led a normal “real life.” The remaining 4 patients (2 missing data) had often thought about it, and 3 of them were worried about the research, expressing concerns about the illness and for their children.

When patients were asked about the meaning of carrying a cancer-predisposing mutation, 8 patients (35%) spontaneously evoked for themselves the awareness of the risk of developing another cancer. Comments from some patients included: “Cancer occurs again; I thought it was all finished,” “My life is getting shorter,” and “I’m not cured; that means relapse, death, and horror.” For their families, 9 patients (40%) evoked concerns about their children that they may carry the mutated gene or they may develop a cancer. Conversely, 6 patients (26%) spoke about the opportunity of suitable medical care for themselves and/or their families. In response to the closed questions, all patients except 1 (96%) thought that they could develop another cancer (“a little” for 14, “a lot” for 4, and “very much” for 4), and 16 patients (70%) said that they were worried about the risk of another cancer (“a little” for 7, “a lot” for 5, and “very much” for 4) versus 7 patients who said that they were not at all worried. Most of the patients who had children (n = 18) said they were worried about their children’s future, and 6 of the 8 patients who wanted children before genetic test results said they were reconsidering their plans to have a family. Five of them explained that they would defer childbearing decisions, and another said they wanted no more children. The patients also expressed concerns about the future atmosphere within their family (n = 5), life assurance or loans (n = 11; 3 missing data), and employment (n = 2; 4 missing data).

The advantages and disadvantages of knowing their genetic status, reported spontaneously by the patients, are summarized in Table 3. A total of 16 patients (70%) said they felt that there were some advantages of knowing their genetic status, especially because they could have suitable medical care. But 17 patients (74%) found also there were disadvantages, mainly because of the stress or worry of cancer (n = 9), and some of them (n = 4) because of the transmission of the mutated gene to their children. In addition, 8 of 19 patients (4 missing data) said that the disadvantages of knowing their cancer-related genetic status outweighed the advantages, and 5 others said that there were as many advantages as disadvantages. However, 9 of these 13 patients said they were aware of this when they had accepted to undergo the blood test. Furthermore, 15 patients said that they would recommend genetic testing to anyone for whom a hereditary cancer was suspected, and all but 1 patient did not regret having decided to undergo genetic testing. The justifications they spontaneously gave were mainly because they wanted to know for themselves and their children (10 patients), to help research to progress for future generations (10 patients), or because it could be useful for better medical care and cancer prevention (5 patients).

One month after result disclosure, the mean score of HADS was 12 (range: 2–20; Table 4). One-third of the patients (n = 7) had a HADS total score of >13, essentially indicating a high anxiety (6 of them had an anxiety score of between 11 and 13). The highest HADS score observed was 20 (Anxiety = 10 and Depression = 10), and this was for 1 patient who was a 38-year-old man who had developed rectal cancer first at 34 and then colon cancer at 37. He was extremely concerned about the risk of relapse and the necessity of a major intervention (colostomy bag). Another high score (=18; Anxiety = 13 and Depression = 05) was for a 53-year-old patient who had developed an early stage colon cancer 5 years ago. He was surprised and worried at result disclosure, stating, “That’s funny! Story finished 5 years ago; I thought it was finished. It could not come back.” He expressed concerns about developing another cancer. The lowest HADS score was 2 for a 33-year-old woman who had developed locally advanced colon cancer at 31 and had relapsed 1 year after, requiring i.p. chemotherapy. Nevertheless, she expressed no emotional changes and no distress reactions at result disclosure (she was then in remission of her disease), explaining she thought she was a carrier.

Despite our small sample, we examined if reactions of the patients to disclosure of their results were influenced by characteristics, such as sex, current age at study, type of hereditary syndrome, and time since diagnosis. Patients with HBOC syndrome and diagnosed with cancer within the previous 5 years more often experienced negative reactions (86 versus 62% and 75 versus 57%). These differences were not statistically significant.

During the month after result disclosure, all of the patients had informed at least one close family member (Table 5). The informed relatives were mainly a partner (83%) and/or siblings (78%) and, less often, children (52%) and parents (30%), because they were considered to be too young or too old to be informed. The patients with children >18 years old (n = 12) had informed at least one of their children, whereas none of the patients who had children under 18 years old (n = 10) had informed them. Half of the 14 patients with at least one parent still alive had informed them; the others argued they did not want to upset them unnecessarily.

Information about genetic testing had been transmitted previously, because all but 1 patient had talked to their close relatives (first degree) and/or spouses about a potential hereditary cancer risk and the blood sample that had been taken. All relatives, once informed, agreed with the decision of the case index patient to undergo molecular research.

Fourteen patients (61%) appreciated or agreed with the confidentiality during result disclosure, and all but 1 said they did not want someone else to have to inform their family. However, 9 of the patients (39%) found some disadvantages to be the only person informed about the positive results, and 6 patients (26%) spontaneously expressed difficulties about being the only person who could transmit the results and genetic information to their family. They mainly reported that they were apprehensive about the potentially harmful nature of the information they had to transmit and the accurate transmission of the technical aspects. Only 2 patients spontaneously reported that they felt guilty, but 9 patients (39%) reported that they felt it was a heavy responsibility to transmit their results.

One month after result disclosure, none of the patients reported conflicts in their families, but 5 patients expressed concerns about the future atmosphere in their families, both for relationships with their partners and with other relatives. One woman was afraid of a possible disagreement with her husband concerning prevention strategy (she was 35 years old and had developed breast cancer 4 years ago); a man expressed concerns about his relationship with his spouse, who seemed to be frail; another woman was afraid of the reactions of her children when she informs them about her results; another woman compared her genetic disorder with a contagious disease; and another expressed concerns about the fact that her relatives will not feel sorry for her.

We have focused on a relatively neglected topic: the impact of a positive genetic test result for cancer patients. Commonly, geneticists and oncologists expect cancer patients to be able to cope well with the disclosure of a positive test result because of their personal history of cancer. Croyle et al. (15) in a prospective study of the psychological responses to BRCA1 mutation test reported among the 25 carriers included that carriers with no cancer history were more distressed than those with cancer history. The authors suggested that the experience of cancer can lead to a reduction of the distress because the patients perceive a greater likelihood that they are mutation carriers. However, Dorval et al. (20) in a subgroup analysis of 17 BRCA1 carriers observed that affected carriers experienced more intense distressed feelings (worry, anger, and/or sadness) to disclosure of a genetic test result than they had anticipated, whereas those unaffected anticipated accurately their emotional reactions. The authors found also that underestimation of anticipation of distressed reactions was associated with an increase in 6-month postdisclosure psychological distress.

Our results confirm that cancer patients could experience distressed reactions to the disclosure of a positive result despite expecting to be a carrier of a cancer-predisposing mutation and reporting a calm period during molecular research. Indeed, all but 2 of the patients said that they were not surprised about their result, and most of them explained that they thought they were a carrier. However, 14 patients reported at least one negative feeling, and besides, 5 patients said they were shocked and/or were discouraged after receiving their results even if this was only transitory. One month after disclosure of the results, the majority of the patients expressed some concerns, mainly about the risk of developing another cancer. “The anguish of cancer comes back,” reported 1 patient; “You no longer feel cured,” reported another, although she had developed cancer a number of years previously. A positive result could disturb the adaptation process, which the patient may have initiated previously to cope with the cancer itself. This can also be illustrated by the example of a man presenting a high HADS score who expressed concerns about the fact that he did not feel cured from his cancer diagnosed 5 years previously. The patient with the lowest HADS score was a woman with an important history of disease and a recent relapse, who said she knew the results of the test and for whom the disclosure did not lead to negative reactions and anxiety. However, patients diagnosed with cancer more recently could also experience reactivation of negative feelings about the cancer diagnosis, as we observed more negative reactions in patients with a time since diagnosis of <5 years compared with the others. Thus, some patients stated, “Cancer occurs again; I thought it was all finished,” and, “My life is getting shorter.” Patients who learn that they are carriers of a cancer-predisposing mutation could begin to feel that their life is very vulnerable. For 1 patient, the positive result represented death. Although the patients knew they would have suitable medical care, they can live in a state of anxiety, fear of relapse, and even death, which could impair their quality of life. In addition, high anxiety may reduce adherence to medical follow-up and care and screening recommendations or prevention strategies (8, 32). It could also be a barrier for the communication of the results to their families. One other concern expressed by most of the patients (∼80%) was their children’s future. Some patients worried that their children may carry the cancer-predisposing mutation and particularly that they would develop the same condition as themselves, i.e., a cancer. They may project their own illness to their offspring.

According to Dorval et al. (20), affected patients, because of their illness, anticipated less accurately their emotional reactions than nonaffected subjects, because it was assumed that they would expect to be a carrier of the mutation and believe that “the worst had already happened.” Moreover, our results suggest that the case index patients could be less prepared than subjects who had undergone predictive testing because of the long delay during molecular research (6–24 months versus 2–4 weeks) and uncertainty of the research, which could be negative and then inconclusive. In addition, being the first subject tested in their family could increase the negative impact of the result disclosure, because carriers may experience more distress as a result of being the first tested in their sibship (19).

Overall, no major psychological adverse effects were observed in our study because the level of emotional distress reflected by the HADS scores was not very high. The mean score was 12, and only 33% patients had a HADS score >13, which was similar to the scores reported in a pilot study evaluating the psychosocial impact of genetic testing in familial medullary-thyroid carcinoma (33) and similar to those observed in a population affected with noninherited cancer (30). However, almost one-third of the patients expressed abnormally high anxiety.

When questioned about genetic testing in general, more than one-third of patients said that the disadvantages of knowing their genetic status outweighed the advantages. However, all patients except 1 did not regret having undergone genetic testing and would recommend it to anyone in the same situation as themselves. Most of them underwent genetic testing essentially to know for themselves and/or their offspring and for scientific advancement, but some did so because they had thought about their own cancer prevention and medical surveillance. Thus, when some patients said that disadvantages outweighed advantages, they probably were talking about the disadvantages of having a hereditary disease that involved themselves and their relatives, rather than of actually knowing their genetic status. Otherwise, they would have regretted having undergone the test, and they would certainly not have recommended it.

We also examined the patients’ behavior about transmission of their results to relatives. As Dudokdewit et al. (34) showed in their description of the first family in the Netherlands for whom predictive testing for HBOC was possible, the case index has to play a crucial role, being “the messenger of the news” within their family. During the month after result disclosure, all of the patients in our study had informed at least one of their close relatives, most often siblings and partners. This suggests that most of the patients will fulfill their role as a messenger within their close family while treating some members cautiously, e.g., parents and young children. Some patients expressed difficulties being the only person who could transmit the test results, although they had talked to their close relatives about genetic testing at the time of research. They dreaded the potentially harmful nature of the information and were afraid that they would not be able to transmit accurately the technical aspects of the results. Additionally, more than one-third of the patients said it was a heavy responsibility. No family conflicts had been reported at the time of this evaluation, but 20% of the patients expressed concerns about the future atmosphere in their family. In addition, 6 patients said they would reconsider their plans to have a child. However, despite the difficulties, all but 1 of the patients in our study did not want someone else to inform their family. This suggests that most of the patients were aware of their important role as “messenger.”

These results should be interpreted with caution given the small size of our sample. However, all of the other published studies have included relatively few affected carrier subjects (15, 20). This low number of affected patients identified as carriers of a BRCA or MMR cancer-predisposing mutation, in the studies, could be explained by the lower than expected frequency of mutation detected in the families at risk. Ford et al. (2) estimated that 59% of families with four or five breast cancers were attributable to genes other than BRCA1 or BRCA2, and Wijnen et al. (35) observed a 26% mutation rate in families suspected of having HNPCC. Furthermore, among the 39 patients who were eligible for the study, 23 (59%) patients participated. Among the patients who did not participate, 12 did not respond to the invitation for the disclosure of their result. Because most of these patients were old (6 were >65 years) and were diagnosed with cancer a long time ago (time since last diagnosis of cancer was ≥12 years for 8), they were less representative of the patients who usually undergo genetic testing. Hence, the impact of the disclosure of a positive test result in specific groups such as elderly cancer patients or patients who had been diagnosed a long time ago should be addressed in future studies.

Finally, we chose to perform our evaluation with semistructured interviews conducted by the same clinical research nurse, which enables the patients’ own opinions to be taken into consideration more than a self-questionnaire with only closed questions. This choice in methodology precludes a multicentric study, which could increase the number of patients recruited. Our sample is made up of two different hereditary cancers, HBOC and HNPCC, but restricted to patients affected with cancer who were counseled and evaluated with similar procedures in a single center. Nevertheless, additional large-scale studies are needed to confirm some differences we have observed concerning the reactions and behavior of the patients according to the type of hereditary cancer predisposition, the time between diagnosis and result disclosure, or other variables, such as sex and age. Multivariate analyses may enable the identification of subjects more likely to experience distressed reactions, and therefore, they will need appropriate support.

In conclusion, our results suggest that the extent of emotional reactions to a positive result for genetic testing in cancer patients should not be underestimated under the pretext that the patients have already coped with a diagnosis of cancer, and therefore, they expect “the worse.” Although patients did not say they regretted having undergone testing, a significant number of them expressed reactions congruent with distress and found difficulties in transmitting their results to relatives. An additional support program could then be envisaged, which may help the patients in their role as messengers within their families. The opportunity to discuss with professionals how they can share results and what potential reactions they can expect from relatives may facilitate the transmission of information. In addition, this support program may improve patients’ wellbeing by helping them to cope with their genetic status, their concerns about their own high risk of developing another cancer, and the risk for their children. It should begin at the time of starting molecular research to prepare them for a positive result disclosure; specially if the probability of detecting a mutation is high.

Despite potential limitations of the results presented here, geneticists and oncologists should bear in mind the potentially negative impact of the disclosure of a positive genetic test result in cancer patients.

Additional large-scale studies are warranted to confirm our findings and to identify the factors that contribute to the potential negative impact of diagnostic genetic testing in patients diagnosed with cancer.

Appendix 1. 

We thank Manon Friedreich who conducted the semistructured interviews. We also thank Prof. Thierry Philip for his helpful advice, as well as Margaret Haugh for her help with the preparation of this manuscript.