Psychological Impact of Genetic Testing for Hereditary Nonpolyposis Colorectal Cancer
http://www.100md.com
《临床肿瘤学》
The University of Texas M.D. Anderson Cancer Center
the University of Texas Health Science Center at Houston School of Public Health, Houston, TX
ABSTRACT
PURPOSE: This study examines the impact of hereditary nonpolyposis colorectal cancer (HNPCC) genetic test results on psychological outcomes among cancer-affected and -unaffected participants up to 1 year after results disclosure.
PATIENTS AND METHODS: A total of 155 persons completed study measures before HNPCC genetic testing, and at 2 weeks and 6 and 12 months after disclosure of test results.
RESULTS: Mean scores on all outcome measures remained stable and within normal limits for cancer-affected participants, regardless of mutation status. Among unaffected carriers of HNPCC-predisposing mutations, mean depression, state anxiety, and cancer worries scores increased from baseline to 2 weeks postdisclosure and decreased from 2 weeks to 6 months postdisclosure. Among unaffected noncarriers, mean depression and anxiety scores did not differ, but cancer worries scores decreased during the same time period. Affected and unaffected carriers had higher mean test-specific distress scores at 2 weeks postdisclosure compared with noncarriers in their respective groups; scores decreased for affected carriers and all unaffected participants from 2 weeks to 12 months postdisclosure. Classification of participants into high- versus low-distress clusters using mean scores on baseline psychological measures predicted significantly higher or lower follow-up scores, respectively, on depression, state anxiety, quality of life, and test-specific distress measures, regardless of mutation status.
CONCLUSION: Although HNPCC genetic testing does not result in long-term adverse psychological outcomes, unaffected mutation carriers may experience increased distress during the immediate postdisclosure time period. Furthermore, those with higher levels of baseline mood disturbance, lower quality of life, and lower social support may be at risk for both short- and long-term increased distress.
INTRODUCTION
Hereditary nonpolyposis colorectal cancer (HNPCC) is the most commonly occurring inherited colorectal cancer (CRC) syndrome, accounting for up to 5% of all CRC patient cases diagnosed annually.1–3 The genetic cause for HNPCC is attributed to mutations in DNA mismatch repair genes, and clinical genetic testing can identify HNPCC-predisposing mutations at several loci.4–8 Because mutation carriers have a 60% to 80% lifetime risk of developing CRC as well as an increased risk for extracolonic cancers (eg, endometrial and stomach cancers),3 they are advised to follow high-risk surveillance recommendations.9
Genetic testing should begin with a family member who has been diagnosed with an HNPCC syndrome-related cancer (ie, a proband). If a deleterious mutation is identified, testing can be offered to the proband's family members because they are at risk of carrying the mutation. Testing negative for a known deleterious HNPCC mutation in one's family may bring relief because it indicates that one is not at increased cancer risk. However, if genetic testing fails to find a pathogenic mutation in a proband whose family history is suggestive of HNPCC, this result is considered uninformative and members of such families are advised to adhere to high-risk surveillance recommendations based on their family cancer history alone. For these families, genetic testing does not help determine which relatives may or may not be at increased risk of developing HNPCC-related cancers.
Knowing one's genetic risk for hereditary cancers may facilitate the early detection or prevention of cancer. However, negative psychological sequelae may develop, particularly among individuals who learn that they are personally at increased risk for developing the disease or for passing on cancer-predisposing genes to their offspring.10 Studies that have examined the psychological impact of receiving results from cancer genetic testing have largely focused on carrier testing for BRCA1 and BRCA2 gene mutations.11–15 Carriers consistently showed greater distress compared with noncarriers after disclosure of test results.11–14 By the same token, notification that one is not a carrier may bring relief and reduce one's distress about developing cancer.
Few studies have examined the psychological impact of genetic counseling and testing for hereditary colon cancer. Aktan-Collan16 evaluated psychological outcomes before HNPCC genetic testing and for up to 1 year after disclosure of test results in 271 individuals with no personal cancer history from 36 Finnish families with known HNPCC mutations. Mutation carriers' state anxiety levels increased immediately after disclosure compared with both noncarriers' and their own baseline levels; however, anxiety levels at 1 year postdisclosure were similar to baseline levels for both carriers and noncarriers and did not differ between the two groups.
Less is known about the impact of HNPCC genetic testing on those with a personal history of cancer. Our previous studies suggest that subgroups of CRC patients undergoing HNPCC testing may be at higher risk for psychological distress after disclosure of test results, on the basis of certain baseline characteristics.17,18 Among CRC patients undergoing HNPCC testing, we found higher levels of depressive symptoms and/or anxiety among women, younger persons, and nonwhites, as well as among those with less formal education and fewer and less satisfactory sources of social support.17 Higher baseline levels of psychological distress, as well as lower quality of life (QOL) and fewer and less satisfactory sources of social support, characterized a subgroup of CRC patients from the same population. Persons in this high-distress subgroup also were more likely to worry about finding out that they were HNPCC mutation carriers and about being able to cope with learning their test results,18 suggesting that they may be at risk for adverse responses to the disclosure of genetic test results, especially if the results are perceived as unfavorable.
Here we extend our previous work by reporting longitudinal findings from our prospective study of patients and unaffected relatives who underwent HNPCC genetic testing.17–20 Our purpose was to evaluate the psychological impact of disclosure of HNPCC genetic test results among individuals with and without a personal history of cancer over a 1-year period.
PATIENTS AND METHODS
Study Population
This study was approved by the Institutional Review Board at The University of Texas M.D. Anderson Cancer Center (MDACC). The study population included 155 adults 18 years of age or older who participated in concurrent studies involving HNPCC genetic testing and psychosocial questionnaires.21 Eligible index patients (the first persons tested in a family) were recruited from GI cancer clinics at MDACC, had a diagnosis of colorectal adenocarcinoma, and also had a family cancer history that met the Amsterdam criteria for HNPCC or otherwise suggested an increased risk of carrying an HNPCC-predisposing mutation (eg, CRC diagnosed age 45 years, multiple relatives and/or generations affected with HNPCC-related cancers). Biologic relatives of mutation-positive index patients who were at 50% or 25% risk of carrying a mutation also were eligible. In this analysis, affected persons included any index patients or relatives with a prior diagnosis of any cancer excluding nonmelanoma skin cancer; unaffected participants included relatives with no personal history of cancer.
Study Procedures
In conjunction with a protocol that offered HNPCC genetic counseling and testing at no cost, we conducted a longitudinal study of psychosocial outcomes associated with HNPCC genetic testing. Recruitment, baseline assessment, and genetic testing of probands took place at MDACC, and has been described in detail elsewhere.18,21,22 When test results became available, probands were contacted again to schedule a genetic counseling appointment for results disclosure. A genetic counselor reviewed pedigrees with mutation-positive probands and identified appropriate family members for genetic counseling and testing, and for recruitment to the psychosocial study. Probands were asked to provide permission to contact eligible relatives; the genetic counselor contacted those family members by letter and telephone to provide a basic description of the family's risk for HNPCC, to describe the psychosocial study, and to offer genetic counseling. Relatives who gave informed consent for the psychosocial study were contacted by a research assistant to arrange a telephone interview for completion of the psychosocial questionnaire, and also were offered a genetic counseling appointment. Relatives could have genetic counseling without participating in the psychosocial study. Relatives who underwent genetic testing were notified when their test results became available and were contacted again to schedule a counseling appointment for results disclosure.
Measures
The instrument battery for this study has been described previously.21 After informed consent, all measures were administered at baseline (before genetic testing), and at 2 weeks, 6 months, and 1 year after the disclosure of genetic test results, with the exception of the Revised Impact of Events Scale (RIES). The RIES was administered at the postdisclosure assessment points only.
Independent Variables
Sociodemographic and medical characteristics. We assessed age, sex, ethnicity, education, income, marital status, number of children, and religious preference. Family cancer history and HNPCC mutation status were obtained from genetic counseling records.
Social support. The short form of the Social Support Questionnaire (SSQ)23 was administered at baseline only, and was included as a variable in the cluster analysis described in the following section. The SSQ is a 12-item scale with two dimensions: the number of persons available for support, and satisfaction with that source of support (scores range from 6 to 36). Higher scores indicate both greater availability of and satisfaction with social support.
Dependent Variables
The following measures were included as dependent variables for all respondents.
Psychological mood assessment. Three validated, reliable measures of mood were used. The Center for Epidemiologic Studies Depression (CES-D) Scale24,25 is a 20-item measure that has been widely used in general-population samples. Scores range from 0 to 60, with higher scores indicating a greater frequency and/or intensity of depressive symptoms. A cutoff score of 16 or greater has been used to denote persons with depressive symptomatology. The state anxiety subscale of the State-Trait Anxiety Inventory (STAI-S) is a 40-item scale that measures state or transitory anxiety.26 Scores range from 20 to 80, with higher subscale scores indicating greater anxiety. The 15-item RIES measures distress specific to receiving genetic testing results, and measures the extent to which a person is experiencing signs or symptoms of intruding thoughts or periods of avoidance, blocking, or denial of distress.27 Overall scores range from 0 to 75, with higher scores indicating greater genetic test-specific distress.
QOL. QOL was measured using the Ferrans and Powers Quality of Life Index (QLI). The QLI measures satisfaction across four life domains, including health and functioning, psychological and spiritual, and socioeconomic subscales. Scores range from 0 to 30, with higher scores indicating better QOL.
Additional Dependent Variables
We included two additional dependent variables for unaffected individuals.
Cancer worries. We used a three-item scale that assesses current frequency of worry about cancer, and the impact of that worry on mood and the ability to perform daily activities.28 Scores range from 1 to 4, with higher scores indicating more frequent worries about cancer.
Perceived risk of CRC. Perceived risk of developing CRC was assessed by a single item: "In your opinion, compared with other persons your age, would you say your chances of getting CRC are: 1 = much lower, 2 = a little lower, 3 = about the same, 4 = a little higher, 5 = much higher." This item was adapted from a core set of measures recommended by the Measurement Task Force of the Cancer Genetics Studies Consortium.29
Analysis
The primary outcomes of interest included psychological mood measures (ie, depression, anxiety, genetic testing-specific distress) and QOL. Among unaffected individuals, additional outcomes included cancer worries and perceived risk of developing CRC. A total of 89 affected and 66 unaffected individuals completed study questionnaires at baseline, at 2 weeks, and at 6 and 12 months after disclosure of results, and were included in this analysis. All analyses were performed with SAS PROC MIXED (SAS System for Mixed Models; SAS Institute Inc, Cary, NC). Statistical significance was set at P < .05.
Impact of genetic test results over time. Analytic methods for longitudinal data were used for evaluating the outcomes over time. We compared mean scores on the outcome measures at baseline with those at 2 weeks and at 6 and 12 months after test results disclosure using the general linear mixed model. Participant identification across time was used as the repeating factor, and both genetic test result (mutation carrier v noncarrier) and assessment point were used as fixed factors. To control for potential correlations among members of the same family, family of origin was entered as a random factor in the model for the analyses involving probands' relatives. The outcomes were first regressed over time on age, education, race or ethnicity, and sex. Test result and assessment point were then entered into the model. Demographic variables including age, sex, race or ethnicity, and education were used as covariates to correct for potential baseline imbalances and to reduce the residual error variance. Interactions between time and test result were evaluated to determine if the patterns of change over time differed for those who were mutation carriers versus noncarriers. Posthoc contrasts were evaluated for the interactions of time with test result. Pairwise comparisons were used to highlight outcome differences between baseline and 2 weeks, between 2 weeks and 6 months, and between 6 and 12 months for the two groups. We adjusted for small-sample correction using the Kenward and Roger method.30
Effect of baseline distress on outcomes after genetic test result disclosure. A classification of participants using baseline scores on four psychological measures (CES-D, STAI-S, QLI, and SSQ) was defined with cluster analysis for affected and unaffected subgroups using hierarchical cluster analysis and the k-means cluster algorithm in the Statistical Package for the Social Sciences, version 12 (SPSS; SPSS Inc, Chicago, IL). Two well-defined clusters were identified for each subgroup in terms of the Euclidean distance between the cluster centers.
To evaluate the effect of baseline distress (low v high) on responses to HNPCC genetic counseling and testing, we analyzed the outcome variables by distress cluster at 2 weeks and at 6 and 12 months after results disclosure. At each time point, the outcome variables were evaluated using a random regression model with baseline distress as the main predictor variable, and mutation status and demographic variables as covariates. To control for correlations among members of the same family, the family of origin was entered as a random factor in the model for analyses involving unaffected participants. The effect of baseline distress at each assessment point was evaluated by comparing the mean scores on outcome variables between the high-distress and the low-distress subgroups, adjusting for mutation status, demographic covariates that were significant, and family of origin.
RESULTS
Completion of Genetic Testing and Study Interviews
A total of 126 affected individuals representing 79 families completed the psychosocial questionnaire and provided a blood sample and consent for genetic testing; of these, 15 (12%) did not receive their results. We also invited 178 unaffected relatives of mutation-positive index patients to participate in the study; 68 (38%) refused and 24 (13%) completed the questionnaire only. Of the 86 unaffected participants who completed the questionnaire and gave a blood sample for testing, four (5%) did not receive their results. Individuals who declined their genetic test results did not differ from those who received results on demographic or distress variables. Of 193 individuals who underwent genetic counseling and testing in this study, 155 (80%) completed all of the follow-up interviews and were included in this analysis. Those who did not complete all the interviews did not differ from those who completed them on baseline demographic or mood variables.
Among 89 affected participants, 33 were HNPCC-predisposing mutation carriers. In the remaining 56 affected participants, genetic testing either failed to identify a mutation or identified a variant of unknown significance and was considered uninformative with regard to their family's HNPCC risk status. Thus, in the following analyses, the affected group included those with either positive or uninformative test results. Among unaffected participants, 19 were HNPCC mutation carriers and 47 were noncarriers who received definitive negative test results.
Characteristics of Study Participants
Most (89%) affected individuals had been diagnosed with CRC and were less likely to have a family history of cancer that met the Amsterdam criteria than were unaffected participants (47% v 83%; P < .01). Table 1 lists a comparison of demographic characteristics among affected and unaffected participants.
Longitudinal Analysis of Outcome Measures by Mutation Carrier Status
For affected participants, race or ethnicity and education were significantly associated with mean scores on the CES-D, STAI-S, and RIES over time; nonwhites had significantly higher mean scores on all measures compared with whites (F = 11.34, P < .01; F = 12.05, P < .01; and F = 14.98, P < .001, respectively) and those with lower educational levels had significantly higher scores compared with those who had higher educational levels (F = 14.67, P < .01; F = 4.06, P < .05; and F = 7.49, P < .01, respectively). Table 2 lists the mean scores on four of the psychological outcome variables at pre- and post-testing assessment points for affected individuals by mutation status. At baseline, carriers and participants with uninformative results did not differ significantly on any of the outcomes. The comparisons between baseline and post-test assessment points by mutation status reveal no significant changes in mean scores on the depression, state anxiety, and QOL measures.
Among unaffected participants, race or ethnicity and education were significantly associated with mean CES-D and STAI-S scores over time, with nonwhites having significantly higher mean scores compared with whites (F = 14.12, P < .01; and F = 4.9, P < .05, respectively), and those with lower educational levels having significantly higher scores compared with those who had higher educational levels (F = 12.89, P < .01; and F = 4.06, P < .05, respectively). Nonwhites also had significantly higher mean RIES scores compared with whites (F = 17.47; P < .001).
Table 3 lists the mean scores on psychological variables at pre- and post-testing time points by mutation status for unaffected participants. No significant correlations were observed among members of the same family for any of the outcomes. Unaffected carriers and noncarriers did not differ significantly at baseline on any outcomes. Longitudinally, comparisons revealed significant interactions between time and mutation status on mean depression and state anxiety scores, with carriers showing a significant increase in mean scores on these measures from baseline to 2 weeks compared with noncarriers, whose mean scores did not differ significantly from baseline to 12 months. Among carriers, post hoc comparisons revealed a significant decrease in mean scores for state anxiety and depression from 2 weeks to 6 months postdisclosure. Mean cancer worries scores revealed a time-by-result interaction effect, with carriers showing a significant increase in mean scores from baseline to 2 weeks, whereas noncarriers' scores decreased significantly during that same time period. Mean scores on the measure of perceived risk of CRC also showed a significant time-by-result interaction; scores remained high from 2 weeks to 12 months for mutation carriers and decreased from baseline to postdisclosure time points for noncarriers.
Figure 1 displays the pattern of change in test-specific distress among both affected and unaffected participants during the postdisclosure time period. Mean RIES scores are shown for each group at the postdisclosure assessment points by carrier versus noncarrier status. Affected carriers had significantly higher mean RIES scores at both 2 weeks and 6 months postdisclosure compared with affected participants with uninformative results (P < .01). Affected carriers' scores also decreased significantly from 2 weeks to 12 months postdisclosure (P < .05); however, scores for participants with uninformative results did not change during this time period. Among unaffected participants, carriers had significantly higher mean RIES scores compared with noncarriers at the 2 week and 6 month postdisclosure time points. Post hoc comparisons revealed a significant reduction in mean RIES scores for both unaffected carriers and noncarriers from 2 weeks to 6 months postdisclosure, and for unaffected carriers alone from 2 weeks to 12 months postdisclosure. Mean RIES scores were significantly higher for unaffected carriers compared with affected carriers at 2 weeks postdisclosure, but did not differ at later assessment points.
Effect of Baseline Distress on Postdisclosure Outcomes
As shown in Table 4, cluster analysis identified two well-defined subgroups for affected and unaffected participants using the mean baseline scores on the CES-D, STAI-S, QLI, and SSQ scales. The high-distress cluster included higher mean depression and anxiety scores, and lower scores on the QLI and SSQ; in addition, more participants in this cluster had mean CES-D scores at or above the cutoff of 16 than below it (85% v 16%; P < .001). Conversely, the low-distress cluster included lower mean depression and anxiety scores, higher QLI and SSQ scores, and a greater proportion with CES-D scores below the cutoff. The proportion comprising the high- and low-distress groups was the same among both affected and unaffected participants (30.3%).
Table 5 lists the mean scores on psychological variables postdisclosure by baseline distress cluster for affected individuals, adjusted for mutation status, family of origin, race or ethnicity, and education. Comparisons of the adjusted mean scores between the high-distress and low-distress groups on the outcome variables at 2 weeks after disclosure revealed a significant baseline distress effect on depression, state anxiety, QOL, and test-specific distress, regardless of mutation status. The high-distress group had significantly higher scores on these measures compared with those in the low-distress group, regardless of mutation status, and this effect persisted at 6 and 12 months postdisclosure. Table 6 lists the mean scores on the outcome variables postdisclosure by baseline distress cluster for the unaffected group. We again observed a significant baseline distress effect on all outcomes through the 12-month postdisclosure period. However, cancer worries and perceived risk scores were not affected by distress, and changes in these scores are primarily related to mutation status.
DISCUSSION
This is the first report to prospectively document the long-term psychological responses and patterns of distress subsequent to HNPCC genetic testing among individuals with and without a personal history of cancer. Our study addressed the void of knowledge regarding the impact of HNPCC genetic testing on persons with a personal cancer history. In general, we did not observe adverse effects on psychological outcomes after the disclosure of positive or negative genetic test results for HNPCC among either affected or unaffected participants. This finding is consistent with studies evaluating genetic testing for BRCA1 and BRCA2.11–15 However, unlike studies involving hereditary breast and ovarian cancer syndrome, our sample comprised similar proportions of both women and men. Previous studies that examined genetic counseling and testing for HNPCC have focused primarily on relatives of probands who were at 50% risk of carrying an HNPCC mutation.16,31 In contrast, our study included individuals with a personal history of cancer, which may be more consistent with current clinical practice in HNPCC genetic counseling and testing. Genetic testing is most informative when first performed on affected members of families at risk of HNPCC,32 and affected individuals' responses to their genetic test results may influence their relatives' uptake of testing and psychological outcomes in relation to counseling and testing process.
It is reassuring that among affected participants, our outcome measures did not differ significantly over time or by HNPCC mutation status. The finding that undergoing genetic testing did not seem to increase the distress burden of cancer patients, regardless of carrier status, could be shared with patients who are considering genetic testing to ease their concerns about possible reactions. It also could be shared with clinicians to encourage them to offer testing to cancer patients when appropriate, especially because these index patients are important in reaching other at-risk family members.
Unaffected carriers did experience a transient, significant increase in mean depression, anxiety, and cancer worries scores immediately after disclosure of their mutation-positive status, but their distress declined over the 1-year observation period. Perceived risk of developing CRC also peaked after disclosure among mutation-positive unaffected participants, and remained slightly elevated during the year, compared with the group with negative results, in which the scores declined from baseline during the follow-up period. It is not surprising that unaffected carriers experienced the greatest psychological impact compared with other participants at 2 weeks after disclosure of test results because this group must adjust to the notification that they are at increased risk of developing cancer. Here too, both persons undergoing testing and clinicians should be aware that some mood disturbance after disclosure is an appropriate reaction, and will likely decline over time. However, if negative mood states persist, these persons should be informed of counseling resources or referred directly.
Prior research from our group showed a clustering of responses from the mood (anxiety and depression), QOL, and social support scales into high- and low-distress subgroups among CRC patients.18 This report substantiates that baseline distress is predictive of postdisclosure distress at all three follow-up points in both affected and unaffected participants. This finding held for the two mood measures, anxiety and depression, and for QOL as well as the RIES. Further reinforcing the stability of the high-distress versus low-distress groups, the baseline scores of the unaffected and affected high-distress groups were similar to one another, as were the low-distress subgroups, at each follow-up point on mood and QOL scales after adjusting for mutation status, family of origin, race or ethnicity, and education. A difference does appear on the RIES, in which scores of unaffected participants decline to lower levels at the longer term follow-ups (6 and 12 months) compared with affected participants, regardless of distress group. This finding may indicate a greater sense of control about screening for and/or preventing disease onset compared with individuals living with cancer.
Other studies have found similar differences in post-test distress between carriers and noncarriers who underwent BRCA1/BRCA2 genetic testing14,15; however, in many cases those differences were due to noncarriers' reduction in distress immediately after disclosure of negative results, rather than to an increase in distress after disclosure of positive results. In contrast, postdisclosure levels of depression, state anxiety, and cancer worries observed among unaffected carriers in our study suggest that positive genetic test results may have a negative short-term impact on mood in this subgroup. Our findings may accurately reflect the initial reactions and subsequent adjustment of unaffected persons who are notified that their HNPCC mutation status substantially increases their lifetime cancer risk. Indeed, an appropriately increased level of concern about cancer risk may motivate the adoption of screening recommendations, and thus secondary prevention of colorectal and other cancers in the HNPCC syndrome. We also expected that having a personal history of cancer would affect responses to disclosure of genetic test results; namely, positive gene tests would be less distressing to those with a personal history of cancer compared with those without such a history. Our findings support this view; however, we found that baseline distress strongly predicted both subsequent general and test-specific distress and QOL, independent of mutation status.
Several limitations of the study exist. Probands were recruited and tested before the wide use of microsatellite instability or immunohistochemistry tumor testing as a prerequisite to gene sequencing; thus, these additional criteria for determining the likelihood of carrying an HNPCC mutation were not used in this study. However, other risk assessment criteria to determine suitability for HNPCC genetic testing were consistent with those used in current clinical practice, and we do not believe any bias was introduced into participant selection. We had incomplete follow-up data on some participants who underwent genetic testing; however, respondents and nonrespondents did not differ on demographic and mood variables, suggesting again that selection bias was unlikely. Finally, because our sample was largely white and well educated, our findings may not generalize to other populations. Conversely, because our study offered genetic counseling and testing at no cost to participants, cost was eliminated as a barrier to counseling and testing, thus enabling us to obtain a more representative sample from the families whom we recruited for this study.
The disclosure of uninformative genetic test results to 63% of affected participants did not negatively affect their mood, despite the fact that they were unable to obtain information from genetic testing that would offer definitive or reassuring information regarding their family's risk for HNPCC. Post-testing distress profiles of persons who received uninformative results were similar to those of mutation carriers, suggesting they continued to perceive themselves at elevated risk. This is an important finding because a limitation of HNPCC genetic testing is the possibility that one may receive an uninformative result. Those persons may benefit from additional counseling and support to ensure that they fully comprehend their test results, that they (alternatively) do not misconstrue the meaning of their results as a true negative, and that they are able to accurately convey this information to their family members. Continued follow-up by genetic counselors or physicians would ensure that they are informed of emerging technologies for additional evaluation of their mutation status.
Our study shows that HNPCC genetic testing may not result in long-term adverse psychological outcomes for many persons with or without a personal history of cancer; however, subgroups of individuals who undergo testing may be at risk for increased distress in the period immediately after results disclosure as well as over time. Persons without a personal history of cancer may be particularly vulnerable to increased psychological distress immediately after notification that they carry an HNPCC-predisposing mutation, and may benefit from additional counseling and support to facilitate their adjustment to this notification. Our study is the first to use a composite distress measure that delineates subgroups on the basis of high and low distress before any genetic counseling and testing. To ensure a rigorous methodology, we used several well-validated measures to evaluate psychological distress; however, additional work to identify a brief, reliable tool to screen for distress before genetic counseling would be beneficial to clinicians.
Our findings also suggest that both affected and unaffected persons in the high-distress subgroup may benefit from more intensive initial counseling as well as follow-up over time. Additional research should address whether assessing or screening for such high-distress individuals and providing an intervention to facilitate coping and adjustment will alleviate their mood dysfunction. Such counseling may relate to psychological status and adjustment in general, and to the likelihood of undertaking screening and other preventive measures, as well as to communicating to family members about HNPCC risk. Additional research, including longer term follow-up, is needed to determine the extent to which distress status affects behavioral, psychosocial, and family communication variables.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank Jill Sawyer, Patricia Ward, and Wendy Kohlmann, who provided genetic counseling for study participants, and Sapna Kapoor, Linda Solomon, Mary Fitzgerald, and Neel Mann, who assisted with data collection. We also thank the participants for their valuable contributions to our study.
NOTES
Supported by grant No. R01HG01200, National Human Genome Institute, National Institutes of Health, Bethesda, MD, to E.R.G., principal investigator.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Lynch HT, de la Chapelle A: Hereditary colorectal cancer. N Engl J Med 348:919-932, 2003
Aarnio M, Mecklin JP, Aaltonen LA, et al: Life-time risk of different cancers in hereditary non-polyposis colorectal cancer (HNPCC) syndrome. Int J Cancer 64:430-433, 1995
Aarnio M, Sankila R, Pukkala E, et al: Cancer risk in mutation carriers of DNA-mismatch-repair genes. Int J Cancer 81:214-218, 1999
Nicolaides NC, Papadopoulos N, Liu B, et al: Mutations of two PMS homologues in hereditary nonpolyposis colon cancer. Nature 371:75-80, 1994
Palombo F, Gallinari P, Iaccarino I, et al: GTBP, a 160-kilodalton protein essential for mismatch-binding activity in human cells. Science 268:1912-1914, 1995
Papadopoulous N, Nicolaides NC, Wei YF, et al: Mutation of a mutL homolog in hereditary colon cancer. Science 263:1625-1629, 1994
Fishel R, Lescoe MK, Rao MR, et al: The human mutator gene homolog MSH2 and its association with hereditary nonpolyposis colon cancer. Cell 75:1027-1038, 1993
Bronner CE, Baker SM, Morrison PT, et al: Mutation in the DNA mismatch repair gene homologue hMLH1 is associated with hereditary non-polyposis colon cancer. Nature 368:258-261, 1994
Burke W, Petersen G, Lynch P, et al: Recommendations for follow-up care of individuals with an inherited predisposition to cancer: I. Hereditary nonpolyposis colon cancer—Cancer Genetics Studies Consortium. JAMA 277:915-919, 1997
Lerman C: Translational behavioral research in cancer genetics. Prevent Med 26:S65-S69, 1997
Croyle RT, Smith KR, Botkin JR, et al: Psychological responses to BRCA1 mutation testing: Preliminary findings. Health Psychol 16:63-72, 1997
Lerman C, Narods S, Schulman K, et al: BRCA1 testing in families with hereditary breast-ovarian cancer. JAMA 275:1885-1892, 1996
Lerman C, Hughes C, Lemon ST, et al: What you don't know can hurt you: Adverse psychologic effects in members of BRCA1-linked and BRCA2-linked families who decline genetic testing. J Clin Oncol 16:1650-1654, 1998
Lodder L, Frets PG, Trijsburg RW, et al: Psychological impact of receiving a BRCA1/BRCA2 test result. Am J Med Genet 98:15-24, 2001
Schwartz MD, Peshkin BN, Hughes C, et al: Impact of BRCA1/BRCA2 mutation testing on psychologic distress in a clinic-based sample. J Clin Oncol 20:514-520, 2002
Aktan-Collan K, Haukkala A, Mecklin JP, et al: Psychological consequences of predictive genetic testing for hereditary non-polyposis colorectal cancer (HNPCC): A prospective follow-up study. Int J Cancer 93:608-611, 2001
Vernon SW, Gritz ER, Peterson SK, et al: Correlates of psychologic distress in colorectal cancer patients undergoing genetic testing for hereditary colon cancer. Health Psychol 16:73-86, 1997
Gritz ER, Vernon SW, Peterson SK, et al: Distress in the cancer patient and its association with genetic testing and counseling for hereditary non-polyposis colon cancer. Cancer Res Ther Control 8:35-49, 1999
Vernon SW, Gritz ER, Peterson SK, et al: Design and methodology of a study of psychosocial aspects of genetic testing for hereditary colorectal cancer. Ann N Y Acad Sci 833:190-194, 1997
Vernon SW, Gritz ER, Peterson SK, et al: Intention to learn results of genetic testing for hereditary colon cancer. Cancer Epidemiol Biomarkers Prev 8:353-360, 1999 (4 pt 2)
Vernon SW: Participation in colorectal cancer screening: A review. J Natl Cancer Inst 89:1406-1422, 1997
Vernon SW: Risk perception and risk communication for cancer screening behaviors: A review. J Natl Cancer Inst Monogr 25:101-119, 1999
Sarason IG, Levine H, Basham R, et al: Assessing social support: The social support questionnaire. J Pers Soc Psychol 1:127-139, 1983
Radloff LS: The CES-D scale: A self-report depression scale for research in the general population. Appl Psychol Measure 1:385-401, 1977
Weissman MM, Sholomskas D, Pottenger M, et al: Assessing depressive symptoms in five psychiatric populations: A validation study. Am J Epidemiol 106:203-214, 1977
Spielberger CD, Gorsuch RL, Lushene RE, et al: State-Trait Anxiety Inventory Manual. Palo Alto, CA, Consulting Psychologists Press, 1970
Horowitz M, Wilner N, Alvarez W: Impact of event scale: A measure of subjective stress. Psychosom Med 41:209-218, 1979
Lerman C, Trock B, Rimer BK, et al: Psychological and behavioral implications of abnormal mammograms. Ann Intern Med 114:657-661, 1991
Bowen DJ, Patenaude AF, Vernon SW: Psychosocial issues in cancer genetics: From the laboratory to the public. Cancer Epidemiol Biomarkers Prev 8:326-328, 1999
Kenward MG, Roger JH: Small sample inference for fixed effects from restricted maximum likelihood. Biometrics 53:983-997, 1997
Hadley DW, Jenkins J, Dimond E, et al: Genetic counseling and testing in families with hereditary nonpolyposis colorectal cancer. Arch Intern Med 163:573-582, 2003
ASCO Working Group on Genetic Testing for Cancer Susceptibility: American Society of Clinical Oncology policy statement update: Genetic testing for cancer susceptibility. J Clin Oncol 21:2397-2406, 2003(Ellen R. Gritz, Susan K. )
the University of Texas Health Science Center at Houston School of Public Health, Houston, TX
ABSTRACT
PURPOSE: This study examines the impact of hereditary nonpolyposis colorectal cancer (HNPCC) genetic test results on psychological outcomes among cancer-affected and -unaffected participants up to 1 year after results disclosure.
PATIENTS AND METHODS: A total of 155 persons completed study measures before HNPCC genetic testing, and at 2 weeks and 6 and 12 months after disclosure of test results.
RESULTS: Mean scores on all outcome measures remained stable and within normal limits for cancer-affected participants, regardless of mutation status. Among unaffected carriers of HNPCC-predisposing mutations, mean depression, state anxiety, and cancer worries scores increased from baseline to 2 weeks postdisclosure and decreased from 2 weeks to 6 months postdisclosure. Among unaffected noncarriers, mean depression and anxiety scores did not differ, but cancer worries scores decreased during the same time period. Affected and unaffected carriers had higher mean test-specific distress scores at 2 weeks postdisclosure compared with noncarriers in their respective groups; scores decreased for affected carriers and all unaffected participants from 2 weeks to 12 months postdisclosure. Classification of participants into high- versus low-distress clusters using mean scores on baseline psychological measures predicted significantly higher or lower follow-up scores, respectively, on depression, state anxiety, quality of life, and test-specific distress measures, regardless of mutation status.
CONCLUSION: Although HNPCC genetic testing does not result in long-term adverse psychological outcomes, unaffected mutation carriers may experience increased distress during the immediate postdisclosure time period. Furthermore, those with higher levels of baseline mood disturbance, lower quality of life, and lower social support may be at risk for both short- and long-term increased distress.
INTRODUCTION
Hereditary nonpolyposis colorectal cancer (HNPCC) is the most commonly occurring inherited colorectal cancer (CRC) syndrome, accounting for up to 5% of all CRC patient cases diagnosed annually.1–3 The genetic cause for HNPCC is attributed to mutations in DNA mismatch repair genes, and clinical genetic testing can identify HNPCC-predisposing mutations at several loci.4–8 Because mutation carriers have a 60% to 80% lifetime risk of developing CRC as well as an increased risk for extracolonic cancers (eg, endometrial and stomach cancers),3 they are advised to follow high-risk surveillance recommendations.9
Genetic testing should begin with a family member who has been diagnosed with an HNPCC syndrome-related cancer (ie, a proband). If a deleterious mutation is identified, testing can be offered to the proband's family members because they are at risk of carrying the mutation. Testing negative for a known deleterious HNPCC mutation in one's family may bring relief because it indicates that one is not at increased cancer risk. However, if genetic testing fails to find a pathogenic mutation in a proband whose family history is suggestive of HNPCC, this result is considered uninformative and members of such families are advised to adhere to high-risk surveillance recommendations based on their family cancer history alone. For these families, genetic testing does not help determine which relatives may or may not be at increased risk of developing HNPCC-related cancers.
Knowing one's genetic risk for hereditary cancers may facilitate the early detection or prevention of cancer. However, negative psychological sequelae may develop, particularly among individuals who learn that they are personally at increased risk for developing the disease or for passing on cancer-predisposing genes to their offspring.10 Studies that have examined the psychological impact of receiving results from cancer genetic testing have largely focused on carrier testing for BRCA1 and BRCA2 gene mutations.11–15 Carriers consistently showed greater distress compared with noncarriers after disclosure of test results.11–14 By the same token, notification that one is not a carrier may bring relief and reduce one's distress about developing cancer.
Few studies have examined the psychological impact of genetic counseling and testing for hereditary colon cancer. Aktan-Collan16 evaluated psychological outcomes before HNPCC genetic testing and for up to 1 year after disclosure of test results in 271 individuals with no personal cancer history from 36 Finnish families with known HNPCC mutations. Mutation carriers' state anxiety levels increased immediately after disclosure compared with both noncarriers' and their own baseline levels; however, anxiety levels at 1 year postdisclosure were similar to baseline levels for both carriers and noncarriers and did not differ between the two groups.
Less is known about the impact of HNPCC genetic testing on those with a personal history of cancer. Our previous studies suggest that subgroups of CRC patients undergoing HNPCC testing may be at higher risk for psychological distress after disclosure of test results, on the basis of certain baseline characteristics.17,18 Among CRC patients undergoing HNPCC testing, we found higher levels of depressive symptoms and/or anxiety among women, younger persons, and nonwhites, as well as among those with less formal education and fewer and less satisfactory sources of social support.17 Higher baseline levels of psychological distress, as well as lower quality of life (QOL) and fewer and less satisfactory sources of social support, characterized a subgroup of CRC patients from the same population. Persons in this high-distress subgroup also were more likely to worry about finding out that they were HNPCC mutation carriers and about being able to cope with learning their test results,18 suggesting that they may be at risk for adverse responses to the disclosure of genetic test results, especially if the results are perceived as unfavorable.
Here we extend our previous work by reporting longitudinal findings from our prospective study of patients and unaffected relatives who underwent HNPCC genetic testing.17–20 Our purpose was to evaluate the psychological impact of disclosure of HNPCC genetic test results among individuals with and without a personal history of cancer over a 1-year period.
PATIENTS AND METHODS
Study Population
This study was approved by the Institutional Review Board at The University of Texas M.D. Anderson Cancer Center (MDACC). The study population included 155 adults 18 years of age or older who participated in concurrent studies involving HNPCC genetic testing and psychosocial questionnaires.21 Eligible index patients (the first persons tested in a family) were recruited from GI cancer clinics at MDACC, had a diagnosis of colorectal adenocarcinoma, and also had a family cancer history that met the Amsterdam criteria for HNPCC or otherwise suggested an increased risk of carrying an HNPCC-predisposing mutation (eg, CRC diagnosed age 45 years, multiple relatives and/or generations affected with HNPCC-related cancers). Biologic relatives of mutation-positive index patients who were at 50% or 25% risk of carrying a mutation also were eligible. In this analysis, affected persons included any index patients or relatives with a prior diagnosis of any cancer excluding nonmelanoma skin cancer; unaffected participants included relatives with no personal history of cancer.
Study Procedures
In conjunction with a protocol that offered HNPCC genetic counseling and testing at no cost, we conducted a longitudinal study of psychosocial outcomes associated with HNPCC genetic testing. Recruitment, baseline assessment, and genetic testing of probands took place at MDACC, and has been described in detail elsewhere.18,21,22 When test results became available, probands were contacted again to schedule a genetic counseling appointment for results disclosure. A genetic counselor reviewed pedigrees with mutation-positive probands and identified appropriate family members for genetic counseling and testing, and for recruitment to the psychosocial study. Probands were asked to provide permission to contact eligible relatives; the genetic counselor contacted those family members by letter and telephone to provide a basic description of the family's risk for HNPCC, to describe the psychosocial study, and to offer genetic counseling. Relatives who gave informed consent for the psychosocial study were contacted by a research assistant to arrange a telephone interview for completion of the psychosocial questionnaire, and also were offered a genetic counseling appointment. Relatives could have genetic counseling without participating in the psychosocial study. Relatives who underwent genetic testing were notified when their test results became available and were contacted again to schedule a counseling appointment for results disclosure.
Measures
The instrument battery for this study has been described previously.21 After informed consent, all measures were administered at baseline (before genetic testing), and at 2 weeks, 6 months, and 1 year after the disclosure of genetic test results, with the exception of the Revised Impact of Events Scale (RIES). The RIES was administered at the postdisclosure assessment points only.
Independent Variables
Sociodemographic and medical characteristics. We assessed age, sex, ethnicity, education, income, marital status, number of children, and religious preference. Family cancer history and HNPCC mutation status were obtained from genetic counseling records.
Social support. The short form of the Social Support Questionnaire (SSQ)23 was administered at baseline only, and was included as a variable in the cluster analysis described in the following section. The SSQ is a 12-item scale with two dimensions: the number of persons available for support, and satisfaction with that source of support (scores range from 6 to 36). Higher scores indicate both greater availability of and satisfaction with social support.
Dependent Variables
The following measures were included as dependent variables for all respondents.
Psychological mood assessment. Three validated, reliable measures of mood were used. The Center for Epidemiologic Studies Depression (CES-D) Scale24,25 is a 20-item measure that has been widely used in general-population samples. Scores range from 0 to 60, with higher scores indicating a greater frequency and/or intensity of depressive symptoms. A cutoff score of 16 or greater has been used to denote persons with depressive symptomatology. The state anxiety subscale of the State-Trait Anxiety Inventory (STAI-S) is a 40-item scale that measures state or transitory anxiety.26 Scores range from 20 to 80, with higher subscale scores indicating greater anxiety. The 15-item RIES measures distress specific to receiving genetic testing results, and measures the extent to which a person is experiencing signs or symptoms of intruding thoughts or periods of avoidance, blocking, or denial of distress.27 Overall scores range from 0 to 75, with higher scores indicating greater genetic test-specific distress.
QOL. QOL was measured using the Ferrans and Powers Quality of Life Index (QLI). The QLI measures satisfaction across four life domains, including health and functioning, psychological and spiritual, and socioeconomic subscales. Scores range from 0 to 30, with higher scores indicating better QOL.
Additional Dependent Variables
We included two additional dependent variables for unaffected individuals.
Cancer worries. We used a three-item scale that assesses current frequency of worry about cancer, and the impact of that worry on mood and the ability to perform daily activities.28 Scores range from 1 to 4, with higher scores indicating more frequent worries about cancer.
Perceived risk of CRC. Perceived risk of developing CRC was assessed by a single item: "In your opinion, compared with other persons your age, would you say your chances of getting CRC are: 1 = much lower, 2 = a little lower, 3 = about the same, 4 = a little higher, 5 = much higher." This item was adapted from a core set of measures recommended by the Measurement Task Force of the Cancer Genetics Studies Consortium.29
Analysis
The primary outcomes of interest included psychological mood measures (ie, depression, anxiety, genetic testing-specific distress) and QOL. Among unaffected individuals, additional outcomes included cancer worries and perceived risk of developing CRC. A total of 89 affected and 66 unaffected individuals completed study questionnaires at baseline, at 2 weeks, and at 6 and 12 months after disclosure of results, and were included in this analysis. All analyses were performed with SAS PROC MIXED (SAS System for Mixed Models; SAS Institute Inc, Cary, NC). Statistical significance was set at P < .05.
Impact of genetic test results over time. Analytic methods for longitudinal data were used for evaluating the outcomes over time. We compared mean scores on the outcome measures at baseline with those at 2 weeks and at 6 and 12 months after test results disclosure using the general linear mixed model. Participant identification across time was used as the repeating factor, and both genetic test result (mutation carrier v noncarrier) and assessment point were used as fixed factors. To control for potential correlations among members of the same family, family of origin was entered as a random factor in the model for the analyses involving probands' relatives. The outcomes were first regressed over time on age, education, race or ethnicity, and sex. Test result and assessment point were then entered into the model. Demographic variables including age, sex, race or ethnicity, and education were used as covariates to correct for potential baseline imbalances and to reduce the residual error variance. Interactions between time and test result were evaluated to determine if the patterns of change over time differed for those who were mutation carriers versus noncarriers. Posthoc contrasts were evaluated for the interactions of time with test result. Pairwise comparisons were used to highlight outcome differences between baseline and 2 weeks, between 2 weeks and 6 months, and between 6 and 12 months for the two groups. We adjusted for small-sample correction using the Kenward and Roger method.30
Effect of baseline distress on outcomes after genetic test result disclosure. A classification of participants using baseline scores on four psychological measures (CES-D, STAI-S, QLI, and SSQ) was defined with cluster analysis for affected and unaffected subgroups using hierarchical cluster analysis and the k-means cluster algorithm in the Statistical Package for the Social Sciences, version 12 (SPSS; SPSS Inc, Chicago, IL). Two well-defined clusters were identified for each subgroup in terms of the Euclidean distance between the cluster centers.
To evaluate the effect of baseline distress (low v high) on responses to HNPCC genetic counseling and testing, we analyzed the outcome variables by distress cluster at 2 weeks and at 6 and 12 months after results disclosure. At each time point, the outcome variables were evaluated using a random regression model with baseline distress as the main predictor variable, and mutation status and demographic variables as covariates. To control for correlations among members of the same family, the family of origin was entered as a random factor in the model for analyses involving unaffected participants. The effect of baseline distress at each assessment point was evaluated by comparing the mean scores on outcome variables between the high-distress and the low-distress subgroups, adjusting for mutation status, demographic covariates that were significant, and family of origin.
RESULTS
Completion of Genetic Testing and Study Interviews
A total of 126 affected individuals representing 79 families completed the psychosocial questionnaire and provided a blood sample and consent for genetic testing; of these, 15 (12%) did not receive their results. We also invited 178 unaffected relatives of mutation-positive index patients to participate in the study; 68 (38%) refused and 24 (13%) completed the questionnaire only. Of the 86 unaffected participants who completed the questionnaire and gave a blood sample for testing, four (5%) did not receive their results. Individuals who declined their genetic test results did not differ from those who received results on demographic or distress variables. Of 193 individuals who underwent genetic counseling and testing in this study, 155 (80%) completed all of the follow-up interviews and were included in this analysis. Those who did not complete all the interviews did not differ from those who completed them on baseline demographic or mood variables.
Among 89 affected participants, 33 were HNPCC-predisposing mutation carriers. In the remaining 56 affected participants, genetic testing either failed to identify a mutation or identified a variant of unknown significance and was considered uninformative with regard to their family's HNPCC risk status. Thus, in the following analyses, the affected group included those with either positive or uninformative test results. Among unaffected participants, 19 were HNPCC mutation carriers and 47 were noncarriers who received definitive negative test results.
Characteristics of Study Participants
Most (89%) affected individuals had been diagnosed with CRC and were less likely to have a family history of cancer that met the Amsterdam criteria than were unaffected participants (47% v 83%; P < .01). Table 1 lists a comparison of demographic characteristics among affected and unaffected participants.
Longitudinal Analysis of Outcome Measures by Mutation Carrier Status
For affected participants, race or ethnicity and education were significantly associated with mean scores on the CES-D, STAI-S, and RIES over time; nonwhites had significantly higher mean scores on all measures compared with whites (F = 11.34, P < .01; F = 12.05, P < .01; and F = 14.98, P < .001, respectively) and those with lower educational levels had significantly higher scores compared with those who had higher educational levels (F = 14.67, P < .01; F = 4.06, P < .05; and F = 7.49, P < .01, respectively). Table 2 lists the mean scores on four of the psychological outcome variables at pre- and post-testing assessment points for affected individuals by mutation status. At baseline, carriers and participants with uninformative results did not differ significantly on any of the outcomes. The comparisons between baseline and post-test assessment points by mutation status reveal no significant changes in mean scores on the depression, state anxiety, and QOL measures.
Among unaffected participants, race or ethnicity and education were significantly associated with mean CES-D and STAI-S scores over time, with nonwhites having significantly higher mean scores compared with whites (F = 14.12, P < .01; and F = 4.9, P < .05, respectively), and those with lower educational levels having significantly higher scores compared with those who had higher educational levels (F = 12.89, P < .01; and F = 4.06, P < .05, respectively). Nonwhites also had significantly higher mean RIES scores compared with whites (F = 17.47; P < .001).
Table 3 lists the mean scores on psychological variables at pre- and post-testing time points by mutation status for unaffected participants. No significant correlations were observed among members of the same family for any of the outcomes. Unaffected carriers and noncarriers did not differ significantly at baseline on any outcomes. Longitudinally, comparisons revealed significant interactions between time and mutation status on mean depression and state anxiety scores, with carriers showing a significant increase in mean scores on these measures from baseline to 2 weeks compared with noncarriers, whose mean scores did not differ significantly from baseline to 12 months. Among carriers, post hoc comparisons revealed a significant decrease in mean scores for state anxiety and depression from 2 weeks to 6 months postdisclosure. Mean cancer worries scores revealed a time-by-result interaction effect, with carriers showing a significant increase in mean scores from baseline to 2 weeks, whereas noncarriers' scores decreased significantly during that same time period. Mean scores on the measure of perceived risk of CRC also showed a significant time-by-result interaction; scores remained high from 2 weeks to 12 months for mutation carriers and decreased from baseline to postdisclosure time points for noncarriers.
Figure 1 displays the pattern of change in test-specific distress among both affected and unaffected participants during the postdisclosure time period. Mean RIES scores are shown for each group at the postdisclosure assessment points by carrier versus noncarrier status. Affected carriers had significantly higher mean RIES scores at both 2 weeks and 6 months postdisclosure compared with affected participants with uninformative results (P < .01). Affected carriers' scores also decreased significantly from 2 weeks to 12 months postdisclosure (P < .05); however, scores for participants with uninformative results did not change during this time period. Among unaffected participants, carriers had significantly higher mean RIES scores compared with noncarriers at the 2 week and 6 month postdisclosure time points. Post hoc comparisons revealed a significant reduction in mean RIES scores for both unaffected carriers and noncarriers from 2 weeks to 6 months postdisclosure, and for unaffected carriers alone from 2 weeks to 12 months postdisclosure. Mean RIES scores were significantly higher for unaffected carriers compared with affected carriers at 2 weeks postdisclosure, but did not differ at later assessment points.
Effect of Baseline Distress on Postdisclosure Outcomes
As shown in Table 4, cluster analysis identified two well-defined subgroups for affected and unaffected participants using the mean baseline scores on the CES-D, STAI-S, QLI, and SSQ scales. The high-distress cluster included higher mean depression and anxiety scores, and lower scores on the QLI and SSQ; in addition, more participants in this cluster had mean CES-D scores at or above the cutoff of 16 than below it (85% v 16%; P < .001). Conversely, the low-distress cluster included lower mean depression and anxiety scores, higher QLI and SSQ scores, and a greater proportion with CES-D scores below the cutoff. The proportion comprising the high- and low-distress groups was the same among both affected and unaffected participants (30.3%).
Table 5 lists the mean scores on psychological variables postdisclosure by baseline distress cluster for affected individuals, adjusted for mutation status, family of origin, race or ethnicity, and education. Comparisons of the adjusted mean scores between the high-distress and low-distress groups on the outcome variables at 2 weeks after disclosure revealed a significant baseline distress effect on depression, state anxiety, QOL, and test-specific distress, regardless of mutation status. The high-distress group had significantly higher scores on these measures compared with those in the low-distress group, regardless of mutation status, and this effect persisted at 6 and 12 months postdisclosure. Table 6 lists the mean scores on the outcome variables postdisclosure by baseline distress cluster for the unaffected group. We again observed a significant baseline distress effect on all outcomes through the 12-month postdisclosure period. However, cancer worries and perceived risk scores were not affected by distress, and changes in these scores are primarily related to mutation status.
DISCUSSION
This is the first report to prospectively document the long-term psychological responses and patterns of distress subsequent to HNPCC genetic testing among individuals with and without a personal history of cancer. Our study addressed the void of knowledge regarding the impact of HNPCC genetic testing on persons with a personal cancer history. In general, we did not observe adverse effects on psychological outcomes after the disclosure of positive or negative genetic test results for HNPCC among either affected or unaffected participants. This finding is consistent with studies evaluating genetic testing for BRCA1 and BRCA2.11–15 However, unlike studies involving hereditary breast and ovarian cancer syndrome, our sample comprised similar proportions of both women and men. Previous studies that examined genetic counseling and testing for HNPCC have focused primarily on relatives of probands who were at 50% risk of carrying an HNPCC mutation.16,31 In contrast, our study included individuals with a personal history of cancer, which may be more consistent with current clinical practice in HNPCC genetic counseling and testing. Genetic testing is most informative when first performed on affected members of families at risk of HNPCC,32 and affected individuals' responses to their genetic test results may influence their relatives' uptake of testing and psychological outcomes in relation to counseling and testing process.
It is reassuring that among affected participants, our outcome measures did not differ significantly over time or by HNPCC mutation status. The finding that undergoing genetic testing did not seem to increase the distress burden of cancer patients, regardless of carrier status, could be shared with patients who are considering genetic testing to ease their concerns about possible reactions. It also could be shared with clinicians to encourage them to offer testing to cancer patients when appropriate, especially because these index patients are important in reaching other at-risk family members.
Unaffected carriers did experience a transient, significant increase in mean depression, anxiety, and cancer worries scores immediately after disclosure of their mutation-positive status, but their distress declined over the 1-year observation period. Perceived risk of developing CRC also peaked after disclosure among mutation-positive unaffected participants, and remained slightly elevated during the year, compared with the group with negative results, in which the scores declined from baseline during the follow-up period. It is not surprising that unaffected carriers experienced the greatest psychological impact compared with other participants at 2 weeks after disclosure of test results because this group must adjust to the notification that they are at increased risk of developing cancer. Here too, both persons undergoing testing and clinicians should be aware that some mood disturbance after disclosure is an appropriate reaction, and will likely decline over time. However, if negative mood states persist, these persons should be informed of counseling resources or referred directly.
Prior research from our group showed a clustering of responses from the mood (anxiety and depression), QOL, and social support scales into high- and low-distress subgroups among CRC patients.18 This report substantiates that baseline distress is predictive of postdisclosure distress at all three follow-up points in both affected and unaffected participants. This finding held for the two mood measures, anxiety and depression, and for QOL as well as the RIES. Further reinforcing the stability of the high-distress versus low-distress groups, the baseline scores of the unaffected and affected high-distress groups were similar to one another, as were the low-distress subgroups, at each follow-up point on mood and QOL scales after adjusting for mutation status, family of origin, race or ethnicity, and education. A difference does appear on the RIES, in which scores of unaffected participants decline to lower levels at the longer term follow-ups (6 and 12 months) compared with affected participants, regardless of distress group. This finding may indicate a greater sense of control about screening for and/or preventing disease onset compared with individuals living with cancer.
Other studies have found similar differences in post-test distress between carriers and noncarriers who underwent BRCA1/BRCA2 genetic testing14,15; however, in many cases those differences were due to noncarriers' reduction in distress immediately after disclosure of negative results, rather than to an increase in distress after disclosure of positive results. In contrast, postdisclosure levels of depression, state anxiety, and cancer worries observed among unaffected carriers in our study suggest that positive genetic test results may have a negative short-term impact on mood in this subgroup. Our findings may accurately reflect the initial reactions and subsequent adjustment of unaffected persons who are notified that their HNPCC mutation status substantially increases their lifetime cancer risk. Indeed, an appropriately increased level of concern about cancer risk may motivate the adoption of screening recommendations, and thus secondary prevention of colorectal and other cancers in the HNPCC syndrome. We also expected that having a personal history of cancer would affect responses to disclosure of genetic test results; namely, positive gene tests would be less distressing to those with a personal history of cancer compared with those without such a history. Our findings support this view; however, we found that baseline distress strongly predicted both subsequent general and test-specific distress and QOL, independent of mutation status.
Several limitations of the study exist. Probands were recruited and tested before the wide use of microsatellite instability or immunohistochemistry tumor testing as a prerequisite to gene sequencing; thus, these additional criteria for determining the likelihood of carrying an HNPCC mutation were not used in this study. However, other risk assessment criteria to determine suitability for HNPCC genetic testing were consistent with those used in current clinical practice, and we do not believe any bias was introduced into participant selection. We had incomplete follow-up data on some participants who underwent genetic testing; however, respondents and nonrespondents did not differ on demographic and mood variables, suggesting again that selection bias was unlikely. Finally, because our sample was largely white and well educated, our findings may not generalize to other populations. Conversely, because our study offered genetic counseling and testing at no cost to participants, cost was eliminated as a barrier to counseling and testing, thus enabling us to obtain a more representative sample from the families whom we recruited for this study.
The disclosure of uninformative genetic test results to 63% of affected participants did not negatively affect their mood, despite the fact that they were unable to obtain information from genetic testing that would offer definitive or reassuring information regarding their family's risk for HNPCC. Post-testing distress profiles of persons who received uninformative results were similar to those of mutation carriers, suggesting they continued to perceive themselves at elevated risk. This is an important finding because a limitation of HNPCC genetic testing is the possibility that one may receive an uninformative result. Those persons may benefit from additional counseling and support to ensure that they fully comprehend their test results, that they (alternatively) do not misconstrue the meaning of their results as a true negative, and that they are able to accurately convey this information to their family members. Continued follow-up by genetic counselors or physicians would ensure that they are informed of emerging technologies for additional evaluation of their mutation status.
Our study shows that HNPCC genetic testing may not result in long-term adverse psychological outcomes for many persons with or without a personal history of cancer; however, subgroups of individuals who undergo testing may be at risk for increased distress in the period immediately after results disclosure as well as over time. Persons without a personal history of cancer may be particularly vulnerable to increased psychological distress immediately after notification that they carry an HNPCC-predisposing mutation, and may benefit from additional counseling and support to facilitate their adjustment to this notification. Our study is the first to use a composite distress measure that delineates subgroups on the basis of high and low distress before any genetic counseling and testing. To ensure a rigorous methodology, we used several well-validated measures to evaluate psychological distress; however, additional work to identify a brief, reliable tool to screen for distress before genetic counseling would be beneficial to clinicians.
Our findings also suggest that both affected and unaffected persons in the high-distress subgroup may benefit from more intensive initial counseling as well as follow-up over time. Additional research should address whether assessing or screening for such high-distress individuals and providing an intervention to facilitate coping and adjustment will alleviate their mood dysfunction. Such counseling may relate to psychological status and adjustment in general, and to the likelihood of undertaking screening and other preventive measures, as well as to communicating to family members about HNPCC risk. Additional research, including longer term follow-up, is needed to determine the extent to which distress status affects behavioral, psychosocial, and family communication variables.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank Jill Sawyer, Patricia Ward, and Wendy Kohlmann, who provided genetic counseling for study participants, and Sapna Kapoor, Linda Solomon, Mary Fitzgerald, and Neel Mann, who assisted with data collection. We also thank the participants for their valuable contributions to our study.
NOTES
Supported by grant No. R01HG01200, National Human Genome Institute, National Institutes of Health, Bethesda, MD, to E.R.G., principal investigator.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Lynch HT, de la Chapelle A: Hereditary colorectal cancer. N Engl J Med 348:919-932, 2003
Aarnio M, Mecklin JP, Aaltonen LA, et al: Life-time risk of different cancers in hereditary non-polyposis colorectal cancer (HNPCC) syndrome. Int J Cancer 64:430-433, 1995
Aarnio M, Sankila R, Pukkala E, et al: Cancer risk in mutation carriers of DNA-mismatch-repair genes. Int J Cancer 81:214-218, 1999
Nicolaides NC, Papadopoulos N, Liu B, et al: Mutations of two PMS homologues in hereditary nonpolyposis colon cancer. Nature 371:75-80, 1994
Palombo F, Gallinari P, Iaccarino I, et al: GTBP, a 160-kilodalton protein essential for mismatch-binding activity in human cells. Science 268:1912-1914, 1995
Papadopoulous N, Nicolaides NC, Wei YF, et al: Mutation of a mutL homolog in hereditary colon cancer. Science 263:1625-1629, 1994
Fishel R, Lescoe MK, Rao MR, et al: The human mutator gene homolog MSH2 and its association with hereditary nonpolyposis colon cancer. Cell 75:1027-1038, 1993
Bronner CE, Baker SM, Morrison PT, et al: Mutation in the DNA mismatch repair gene homologue hMLH1 is associated with hereditary non-polyposis colon cancer. Nature 368:258-261, 1994
Burke W, Petersen G, Lynch P, et al: Recommendations for follow-up care of individuals with an inherited predisposition to cancer: I. Hereditary nonpolyposis colon cancer—Cancer Genetics Studies Consortium. JAMA 277:915-919, 1997
Lerman C: Translational behavioral research in cancer genetics. Prevent Med 26:S65-S69, 1997
Croyle RT, Smith KR, Botkin JR, et al: Psychological responses to BRCA1 mutation testing: Preliminary findings. Health Psychol 16:63-72, 1997
Lerman C, Narods S, Schulman K, et al: BRCA1 testing in families with hereditary breast-ovarian cancer. JAMA 275:1885-1892, 1996
Lerman C, Hughes C, Lemon ST, et al: What you don't know can hurt you: Adverse psychologic effects in members of BRCA1-linked and BRCA2-linked families who decline genetic testing. J Clin Oncol 16:1650-1654, 1998
Lodder L, Frets PG, Trijsburg RW, et al: Psychological impact of receiving a BRCA1/BRCA2 test result. Am J Med Genet 98:15-24, 2001
Schwartz MD, Peshkin BN, Hughes C, et al: Impact of BRCA1/BRCA2 mutation testing on psychologic distress in a clinic-based sample. J Clin Oncol 20:514-520, 2002
Aktan-Collan K, Haukkala A, Mecklin JP, et al: Psychological consequences of predictive genetic testing for hereditary non-polyposis colorectal cancer (HNPCC): A prospective follow-up study. Int J Cancer 93:608-611, 2001
Vernon SW, Gritz ER, Peterson SK, et al: Correlates of psychologic distress in colorectal cancer patients undergoing genetic testing for hereditary colon cancer. Health Psychol 16:73-86, 1997
Gritz ER, Vernon SW, Peterson SK, et al: Distress in the cancer patient and its association with genetic testing and counseling for hereditary non-polyposis colon cancer. Cancer Res Ther Control 8:35-49, 1999
Vernon SW, Gritz ER, Peterson SK, et al: Design and methodology of a study of psychosocial aspects of genetic testing for hereditary colorectal cancer. Ann N Y Acad Sci 833:190-194, 1997
Vernon SW, Gritz ER, Peterson SK, et al: Intention to learn results of genetic testing for hereditary colon cancer. Cancer Epidemiol Biomarkers Prev 8:353-360, 1999 (4 pt 2)
Vernon SW: Participation in colorectal cancer screening: A review. J Natl Cancer Inst 89:1406-1422, 1997
Vernon SW: Risk perception and risk communication for cancer screening behaviors: A review. J Natl Cancer Inst Monogr 25:101-119, 1999
Sarason IG, Levine H, Basham R, et al: Assessing social support: The social support questionnaire. J Pers Soc Psychol 1:127-139, 1983
Radloff LS: The CES-D scale: A self-report depression scale for research in the general population. Appl Psychol Measure 1:385-401, 1977
Weissman MM, Sholomskas D, Pottenger M, et al: Assessing depressive symptoms in five psychiatric populations: A validation study. Am J Epidemiol 106:203-214, 1977
Spielberger CD, Gorsuch RL, Lushene RE, et al: State-Trait Anxiety Inventory Manual. Palo Alto, CA, Consulting Psychologists Press, 1970
Horowitz M, Wilner N, Alvarez W: Impact of event scale: A measure of subjective stress. Psychosom Med 41:209-218, 1979
Lerman C, Trock B, Rimer BK, et al: Psychological and behavioral implications of abnormal mammograms. Ann Intern Med 114:657-661, 1991
Bowen DJ, Patenaude AF, Vernon SW: Psychosocial issues in cancer genetics: From the laboratory to the public. Cancer Epidemiol Biomarkers Prev 8:326-328, 1999
Kenward MG, Roger JH: Small sample inference for fixed effects from restricted maximum likelihood. Biometrics 53:983-997, 1997
Hadley DW, Jenkins J, Dimond E, et al: Genetic counseling and testing in families with hereditary nonpolyposis colorectal cancer. Arch Intern Med 163:573-582, 2003
ASCO Working Group on Genetic Testing for Cancer Susceptibility: American Society of Clinical Oncology policy statement update: Genetic testing for cancer susceptibility. J Clin Oncol 21:2397-2406, 2003(Ellen R. Gritz, Susan K. )