Understanding the quality of life (QOL) issues in survivors of cancer: towards the development of an EORTC QOL cancer survivorship questionnaire

With continuing improvement in early detection and treatment, and an aging population, the number of cancer survivors is increasing steadily. This has resulted in a growing interest in evaluating the health-related quality of life (HRQOL) of cancer survivors [1]. Increasingly, clinical trials and comparative effectiveness studies are being designed to include long-term follow-up to assess, in addition to survival, late effects of treatment and HRQOL. In order to integrate HRQOL in such studies, it is essential to develop patient-reported outcome measures (PROMs) that capture the full range of issues relevant to disease-free cancer survivors.

Many of the cancer-related HRQOL questionnaires that are available today, including the European Organisation for Research and Treatment of Cancer core questionnaire (EORTC QLQ-C30) [2] and the Functional Assessment of Cancer Therapy – General (FACT-G) [3], with their supplementary site-specific modules, may not be entirely appropriate for assessing the experiences of disease-free cancer survivors. These questionnaires include items assessing acute and treatment-related symptoms (e.g., vomiting) that are typically no longer relevant in the post-treatment survivorship period. Conversely, they may not adequately address physical and psychosocial health problems particularly relevant to cancer survivors (e.g., fear of recurrence, return to work).

Questionnaires that have been developed specifically for use among (long-term) cancer survivors include the Cancer Problems in Living Scale (CPILS) [4, 5], Impact of Cancer (IOC/IOCv2) [6‐8], Quality of Life in Adult Cancer Survivors (QLACS) [9, 10], Quality of Life Cancer Survivors (QoL-CS) [11] and Satisfaction with Life Domains Scale for Cancer (SLDS-C) [12]. These questionnaires focus primarily on psychosocial aspects of survivorship and pay relatively little attention to assessing chronic physical effects of cancer and its treatment [13]. Additionally, while they assess generic (e.g., fear of recurrence) HRQOL issues relevant to the survivorship period, these questionnaires do not include condition-specific issues which may persist or arise as a long-term consequence of treatment (e.g., genitourinary problems in prostate cancer survivors or lymphedema problems in breast cancer survivors). Also, they often have been based on investigations of a limited number of cancer sites and on survivors living in the United States [4‐8, 12], thus limiting their generalizability to other survivor populations in other countries. Finally, for most of these questionnaires, there is only limited information available about their psychometric properties, or the psychometrics have been based on only a small number of cancer survivor populations [13].

For this reason, the EORTC Quality of Life Group (QLG) has embarked on a project with the primary objective of developing a HRQOL assessment approach that captures the full range of issues relevant to disease-free cancer survivors, both in general and for specific cancer sites. Many definitions of cancer survivorship have been used in the literature [14]. We use the term “cancer survivor” to describe any person who has been diagnosed with cancer who has completed treatment with curative-intent (with the exception of maintenance treatment) and is disease-free (no evidence of active cancer). Since The EORTC QLG’s current portfolio of measures has been primarily designed to assess patients’ HRQOL during treatment and shortly after completion of treatment, it makes sense to begin use of survivorship measures once the acute symptoms of the disease and its treatment have resolved. An important secondary objective of this project is to determine the most appropriate minimum time since end of primary treatment for commencing use of survivorship HRQOL measures.

The conceptual framework we employed for the development of the questionnaire followed the World Health Organization (WHO) definition of health, dating from 1948, as “a state of complete physical, mental, and social well-being and not merely the absence of disease or infirmity”. Combined with the Medical Outcomes study (MOS) framework it delineates three key dimensions of health: physical, mental and social [15]. These three dimensions can be assessed by several types of indicators (see Fig. 1). Some of these indicators reflect primarily one of the three dimensions (e.g., physical functioning) and others reflect two or three dimensions (e.g., fatigue) [15].

In terms of measurement strategy, this project follows the tradition within the EORTC QLG of involving patients from a range of countries, cultures and languages at every step in the developmental process. Additionally, given the goal of addressing both generic and cancer site-specific survivorship issues, the project encompasses a broad range of diagnostic groups. We follow the EORTC QLG’s four- phase process of questionnaire development [16]: 1) generation of relevant HRQOL of issues; 2) conversion of HRQOL issues into a set of questionnaire items; 3) questionnaire pre-testing; and 4) large-scale international field testing. In the current paper we report on the results of the first phase of this project.

Phase I had two aims: 1) to determine the full range of issues relevant to disease-free cancer survivors, both in general and for specific cancer sites; and 2) to determine the most appropriate minimum time since end of primary treatment for commencing use of survivorship HRQOL measures. It consisted of two sub-phases: In phase 1a we generated an exhaustive list of HRQOL issues, drawing upon two primary sources: the literature and cancer survivors. In phase 1b we asked a sample of cancer survivors to rate the QLQ-C30 to determine at which time after treatment completion the acute symptoms and side-effects related to cancer and its treatment are no longer relevant. We considered the diminishing prevalence of these symptoms an indicative of the need to shift to the assessment of longer term survivorship issues. In addition, the survivors participating in phase 1b rated the list of HRQOL issues developed in phase 1a to identify the issues relevant to disease free-survivors. In this phase only quantitative data analyses were applied. The workflow of the study is also presented in Fig. 2.

In this first phase of our cancer survivorship questionnaire development project, we identified 116 generic survivorship issues. Additionally, on average, we identified 26 site-specific survivorship issues per tumor site, which only partially overlapped with the existing EORTC site-specific modules. We also observed that, approximately one year following completion of cancer treatment, most of the acute disease- and treatment-related symptoms have resolved themselves in the large majority of survivors.

Based on these findings, we will move forward with the development of a core survivorship questionnaire for disease-free adult survivors who are at least one year post-treatment. This questionnaire will retain many of the original items and scales from the QLQ-C30, deleting only those items that assess acute symptoms (nausea/vomiting, appetite loss, constipation, and diarrhea). Additional survivorship issues will be added to expand the scope of issues addressed by the questionnaire. To improve the measurement precision of some of the existing scales of the QLQ-C30, we will collaborate with the EORTC CAT team [42] to select the issues of the generic issue list and the items of the EORTC QLG item library assessing these issues. This survivorship questionnaire can be complemented by cancer-site specific survivorship modules based on and adapted from the existing EORTC cancer site-specific modules.

The period after treatment completion is often described by survivors as more difficult than the treatment itself [43]. The end of the phase of transition from being a patient to resuming normal life [44] can be very positive, but also brings with it feelings of uncertainty about the future and fear of cancer recurrence. During this early survivorship period, patients often begin to process the emotions related to the diagnosis, to find meaning in their experience of having had cancer, and to deal with the lingering effects of treatment. The end of this turbulent immediate post-treatment period appears to represent an appropriate starting point for assessing survivorship issues, as both physical and psychosocial health begin to stabilize. This was corroborated by the increase in physical, role, and social functioning and the decline in fatigue observed in our study sample after the first post-treatment year. We did not observe a further decline in the acute symptoms of cancer treatment, as the prevalence of acute symptoms was already low a half year after treatment completion. The chronic side effects of treatment (pain, dyspnea, insomnia, and fatigue) continued to be relevant for all survivor groups into the longer post-treatment phase.

In accordance with the existing cancer survivorship questionnaires [4‐12] our results indicate that feelings of uncertainty about the future, fears related to recurrence of cancer, fears and worries concerning family members, feelings of depression and anger, feelings that others do not understand the impact of cancer, positive impact on social relationships, positive changes in (perception of) life, negative body image, cognitive problems, fatigue, sleeping problems, pain, sexual problems, and dealing with the chronic physical consequences of cancer are all relevant issues for cancer survivors. However, our results also indicate that other issues often included in survivorship questionnaires may be less relevant when rated by a wider range of cancer survivors in an international context. This includes issues related to feelings of guilt, fears related to starting new (romantic) relationships, and feelings of pride about having survived cancer. Also, compared to existing questionnaires, our findings underscore the relevance of assessing issues related to chronic side effects of treatment such as neuropathy and joint pain [4‐12]. Over 30% of the issues were related to physical functioning, including chronic physical effects of cancer and its treatment, like Raynaud symptoms, neuropathy, joint pain, and muscle cramps. These issues receive relatively little attention in the existing cancer survivorship questionnaires. The differences between our findings and the existing survivorship questionnaires may reflect both culture (including differences between the American and European health care systems) and the fact that we included a wider range of diagnostic groups in our study, and placed relatively less emphasis on breast cancer and non-solid cancer survivors, as has typified the developmental phase of other survivorship questionnaires.

Another important finding from our study is that there is not only a fairly large number of condition-specific physical health issues in cancer survivorship, but also differences in the extent to which various psychosocial issues are perceived as relevant by specific cancer diagnostic groups. These differences in perceived relevance of survivorship issues may reflect differences in survival rates between the cancer types, the average age at diagnosis, the nature of the chronic side effects of the various treatments, and whether a cancer diagnosis is sex-specific. For example, glioma and lung cancer survivors reported issues related to the negative impact of cancer on their lives and issues related to struggles with family and friends as being very relevant; lung, lymphoma, and colorectal cancer survivors more frequently rated work-related issues; and bladder and head & neck cancer survivors more frequently endorsed role functioning issues. Most of these site-specific survivorship issues are currently not included in the existing site-specific modules of the EORTC.

The literature consistently shows that younger cancer survivors report a higher impact of their cancer experience on HRQOL [45, 46], including higher levels of distress, than older cancer survivors. This is likely related to the fact that relatively younger survivors are confronted with a life-threatening illness at a time when many are in the midst of forming relationship bonds, starting and raising families, and trying to establish a workable balance between career and family life. During this period of young adulthood, a serious illness such as cancer is less expected, and may therefore be more disruptive. Also, younger cancer patients and survivors may perceive themselves as having more to lose in terms of future perspective, and may have fewer opportunities for peer support (i.e., having contemporaries with whom they can share their common experience of having had cancer). Conversely older survivors have more life experience, which might lead to better coping strategies, and they may face fewer work-related and social demands. This is supported by the findings from our study that younger survivors are more likely than older survivors to rate issues related to having children, financial difficulties, loss of future life plans, and lack of support as being relevant to them. Although some of these issues appear to be more relevant to younger survivors, in the interest of parsimony (i.e., not having to create two versions of a core survivorship questionnaire), we have decided to include them in our consolidated issue list.

A strength of our study is that we included a relatively large number of survivors from 11 cancer diagnosis groups from a total of 14 European countries. This enhances the generalizability of our findings. Also, our strategy with site-specific survivor modules ensures that relevant chronic physical symptoms are included. Furthermore, the retention of the items and scales of the QLQ-C30 that are still relevant for disease-free survivors will ensure continuity in the evaluation of HRQOL over time, from diagnosis through the long-term survivorship phase.

A possible limitation of our work is that those cancer survivors in our sample who were more than 5 years post-treatment were drawn primarily from hospital registries. Many patients who are 5 years or longer post-treatment may no longer be in active follow-up, and those who are may be those with more serious, chronic health problems. This could cause some degree of overestimation of the relevance of various survivorship issues in this subgroup of longer term survivors. Also, our sample was somewhat younger than one might expect based on the median age of the general population at cancer diagnosis (66 years) [47].

We identified 116 generic survivorship issues, and on average, 26 site-specific survivorship issues per tumor site. Compared to existing cancer survivorship questionnaires, our findings underscore the relevance of assessing issues related to chronic physical side effects of treatment such as neuropathy and joint pain in addition to the psychosocial aspects of survivorship.

In the next phase of this project, we will further develop and test the core survivorship questionnaire, and we will also develop survivorship modules for breast, prostate, and colorectal cancer survivors. The choice of these three disease sites was based on the incidence, survival rates, and the number of survivorship studies conducted in these disease sites. In the longer term, we intend to develop survivorship modules for a much broader set of cancer sites. Ultimately, this will yield a comprehensive suite of survivorship questionnaires that will yield both a common data set for comparison of results across tumor sites, and unique information about the survivorship experience of specific groups of cancer survivors.