Assessment of factors that contribute to decreased quality of life in Gynecologic Oncology Group ovarian cancer trials

Recently, von Gruenigen et al. published a study to assess which specific Functional Assessment of Cancer Therapy-Ovarian (FACT-O; a 12-item subscale for ovarian cancer) quality of life (QoL) items are associated with low QoL aspects in women with ovarian cancer during the course of their chemotherapy [1]. In general, published studies do not give information about which specific items may be compromised. This could have a great impact on the total QoL because some of these items are representative of symptoms and harmful side effects that may be treated and ameliorated with therapies. In total, 361 Federation of Gynecology and Obstetrics stage III–IV ovarian cancer patients enrolled in two Gynecologic Oncology Group (GOG) studies (GOG 152 and GOG 172) who underwent primary surgery followed by six cycles of cisplatin and paclitaxel chemotherapy were evaluated [2,3]. GOG 152 measured QoL in a randomized trial of interval secondary cytoreduction in women with advanced ovarian carcinoma who underwent suboptimal debulking. GOG 172 measured QoL in a randomized study of intravenous paclitaxel and cisplatin versus intravenous paclitaxel, intraperitoneal cisplatin and intraperitoneal paclitaxel in women who underwent optimal debulking with stage III epithelial ovarian cancer. A total of 208 patients from GOG 152 who did not undergo interval cytoreduction, and 210 patients entered in the intravenous cisplatin and paclitaxel arm from GOG 172, were evaluated. Patients who received intraperitoneal therapy on GOG 172 and patients who underwent secondary debulking surgery on GOG 152 were not included in the current analysis.

The baseline QoL-FACT-O measures in GOG 152 (taken before chemotherapy cycle 4) and the second QoL-FACT-O measurement from GOG 172 (taken before chemotherapy cycle 4) were analyzed.

Patients were divided into quartiles according to total FACT-O score. Women in the lowest quartile (Q1) were compared with those in the upper three quartiles (Q2–4). This comparison was chosen to expand upon the observation of Wenzel et al. that baseline QoL was predictive for survival and this issue was primarily attributed to the lowest scoring quartile [2]. Comparisons were made separately for the GOG 152 and GOG 172 studies, because the two protocols differed in terms of the percentage of patients submitted to optimal debulking surgery.

Each of the 39 items included in the FACT-O scale were examined for patients in Q1 versus patients in Q2, Q3 and Q4. The proportions of women who selected the two worst response categories for each item were compared between the two groups (Q1 vs Q2–4).

The authors reported that in the physical well-being (PWB) domain, a significantly higher proportion of women in Q1 selected the two worst categories of ‘nausea’, ‘pain’, ‘feeling ill’, ‘lack of energy’ and ‘being bothered by the side effects of treatment’. In the functional well-being (FWB) domain, a significantly higher proportion of women in Q1 selected the two worst categories of ‘being able to work’, ‘finding work fulfilling’, ‘being able to enjoy things’ and ‘being content with the quality of my life’. In the emotional well-being (EWB) domain, a significantly higher proportion of women in Q1 reported ‘feeling nervous, sad, and worrying about dying’ as the most QoL-affecting feeling. In the sexual well-being (SWB) domain, significant differences between patients in Q1 and patients in Q2–4 were ‘feeling close to friends’ for patients on GOG 152 and ‘satisfied with my sex life’ for both protocols. Over half of the patients in Q1 on GOG 152 and GOG 172 chose the lowest categories of ‘being satisfied with my sex life’. Between 50 and 75% of all patients (Q1–4) in both protocols chose the lowest categories of ‘being interested in sex’. The FACT-O items that differed significantly between women in Q1 versus women in Q2–4 on both protocols were: ‘having a good appetite’, ‘I like the appearance of my body’ and ‘I am able to feel like a woman’.

Why is it important to investigate QoL in cancer patients?

In 1948, after the WHO had defined health as “not only the absence of infirmity and disease but also a state of complete physical, mental, and social well being” [4], physicians were reminded that the patient’s health was more than just a corporeal state and could be affected by environmental and social factors.

The first measures of the quality of survival in a clinical trial were reported in 1966 in breast cancer patients after radical mastectomy or limited surgery [5]. The postoperative questionnaire contained not only objective measures such as lymphedema and activity status but also an evaluation of the patient’s attitude. Although the authors did not clearly state how the results of the questionnaire were translated into the measures of attitude, this study represents a pioneering effort to include patients’ subjective opinions in comparing the effects of treatment.

After noting that cancer patients were often distressed by the adverse (but unmeasured) symptomatic effects of radiotherapy and chemotherapy, in 1969 Feinstein et al. called for better methods that would measure the “quality of survival”, at least with respect to a patient’s pain, distress or suffering [6].

As a specific concept, the term ‘QoL’ (rather than quality of survival) entered the medical literature in 1966 [7] in an article on patients receiving hemodialysis, but the first QoL measurement instrument (Priestman and Baum’s 1976 Linear Analogue Self Assessment Scale[8]) only became available 10 years later. Despite this, the standard approach for judging the efficacy of cancer therapeutic agents continued to be the quantity of survival.

In 1985 the US FDA announced the decision to require QoL data as one of the ‘key efficacy parameters’ in clinical trials for new anticancer agents [9]. A working group from the FDA and the National Cancer Institute later recommended that validated QoL instruments should be used for comparing either pre- and post-treatment or treatment versus placebo groups [10].

The evaluation of QoL and performance outcomes in cancer is critical to optimal patient care, comprehensive evaluation of treatment alternatives and the development of informed rehabilitative services and patient education [11–13]. QoL measures have now become a vital part of health outcomes appraisal and an effective way of capturing the personal and social context of patients. For example, the UK National Institute for Health and Clinical Excellence utilizes QoL information for assessing the ‘Cost per Quality Adjusted Life Year’ [101]. In other words, with the valuable insight gained from QoL studies, physicians can now not only add years to one’s life but also life to one’s years [14]. Understanding the complex factors that contribute to overall QoL is critical in order to obtain a comprehensive, multidimensional picture of the patient’s ‘total health-related experience’.

Methods for QoL evaluation

The general types of health-related QoL measures include health status assessment and patient preference assessment. Health status assessments measure different components of a woman’s perceived physical, emotional and social well-being. Generic QoL instruments for patients with cancer, such as the European Organisation for Research and Treatment of Cancer (EORTC) QLQ-C30 [15] and the Functional Assessment of Cancer Therapy General (FACT-G) [16] have been validated and are widely used. Ovarian cancer specific subscales are available for both instruments. The EORTC QLQ-OV28 ovarian cancer module consists of 28 items that assess abdominal symptoms, peripheral neuropathy, other chemotherapy-related side effects, hormonal symptoms, body image, sexual functioning and attitudes towards disease and treatment [17,18]. Similarly, the FACT-G can be combined with FACT-O [19].

Patient preference assessments allow researchers to evaluate how patients decide on trade-offs between quantity of life (e.g., overall survival) and QoL; many patients with ovarian cancer often face clinical decisions with uncertain outcomes that require them to balance quantity of life and QoL issues. Several preference assessment studies have been conducted with ovarian cancer patients [20–22]. These studies assessed women’s preferences for side effects of chemotherapy, which are influenced by whether patients have actually experienced treatment toxicities.

Quality of life during chemotherapy in ovarian cancer patients

Lakusta et al. performed a descriptive study to evaluate the variables that influenced QoL in women with ovarian cancer undergoing chemotherapy treatment [23]; the authors demonstrated that QoL scores were significantly worse for patients receiving cisplatin than for those receiving carboplatin as a consequence of the increased toxicity of cisplatin, especially in term of nausea, vomiting and diarrhea. The authors also reported that specific QoL symptoms (appetite, fatigue and pain) were able to significantly predict the disease outcome at 12 months after starting treatment in 81.3% of patients. Greimel et al. also arrived at the same conclusions and reported better QoL scores (physical, cognitive and role functioning) in a population of ovarian cancer patients receiving carboplatin–paclitaxel (TC) chemotherapy with respect to a control group receiving cisplatin–paclitaxel combination chemotherapy [24]. Although hematological toxicity was worse in the TC arm, overall QoL was better, indicating that these toxicity parameters have little impact on QoL, in contrast to non-hematological toxicities (e.g., nausea, vomiting, loss of appetite and neurotoxicity), which interfere with self-care activities, mobility and QoL to a greater extent.

Carey et al. reported that ovarian cancer patients with better baseline global QoL scores had improved outcomes in terms both of progression-free survival and overall survival and this finding was independent of baseline performance status [25]. The authors suggested that, in some instances, the evaluation of specific symptoms as a component of QoL may be related to prognosis but the study was unable to comment on the impact of specific symptoms on prognosis because this was not part of the analysis.

The importance of identifying the specific parameters of QoL evaluation

Taken together, these data highlight the importance of clearly defining which specific baseline QoL scores change over time because of the effect of chemotherapy on patients’ PWB and FWB.

von Gruenigen et al. examined in detail the different domains of QoL questionnaires in women with ovarian cancer undergoing chemotherapy [1]. In this study, women were divided into quartiles on the basis of their own QoL score. Patients in the lower quartile were compared with those in the upper three quartiles in order to investigate which items were associated with lower QoL score in the lowest quartile. The authors hypothesize that these items should represent the target of clinician’s intervention in order to ameliorate the QoL of patients.

For women who had total QoL scores in the lowest quartile (Q1), von Gruenigen et al. demonstrated significant differences in specific concerns, including ‘pain’, ‘nausea’, ‘feeling ill’ and ‘being bothered by the side effects of treatment’ [1]. This is a considerable conclusion in that several of these items may be amenable to interventions that consequently may increase other aspects of PWB, such as ‘having more energy’ and ‘being able to meet the needs of my family’. Future research also should examine whether interventions aimed at ameliorating pain, nausea, feeling ill and being bothered by side effects of treatment really improve FWB.

In the current study, a higher proportion of women who underwent suboptimal debulking selected the worst categories of feeling nervous (EWB) compared with women who underwent optimal debulking (45 vs 23%); so it appears that suboptimal debulking is a burden not only on overall survival but also on EWB in patients with ovarian cancer. This identifies a potential area of intervention (i.e., counseling) to alleviate some of these feelings and concerns, especially among women who undergo suboptimal debulking.

Sexual problems are prevalent among patients with ovarian cancer. Carmack Taylor et al.[26] and Liavaag et al.[27] reported that women with ovarian cancer who were receiving chemotherapy had the lowest sexual activity. The latter study also indicated that sexually active patients reported significant alterations in pleasure and discomfort. The current study demonstrated that, regardless of whether sexual activity was being assessed in terms of satisfaction (SWB) or interest (FACT-O items), the majority of patients were having difficulty in this area, which could represent another area of medical intervention in order to improve global QoL scores.

In conclusion, women with a low QoL should be evaluated to determine whether clinical interventions are appropriate to ameliorate their QoL and also to increase their long-term outcomes given the predictive value that baseline QoL parameters retain on progression-free and overall survival.

Conclusions

A recent editorial published in the Journal of Clinical Oncology pointed out that it is ‘disappointing’ that despite the fact that thousands of patients have been enrolled in cancer clinical trials with a health-related QoL component, “there are relatively few examples of formal quality-of-life measurement that have influenced individual patient decision making or treatment policies” [28]. While a number of reasons could account for the arguments brought up in the editorial, the present findings indicate that the reason could merely stem from the generally poor level of methodological rigor in the health-related QoL assessment of many previously conducted randomized controlled trials.

In fact, even in the trials that reported QoL measures, the quality of the analysis has not always been adequate. An EORTC study reported that only 39.3% of QoL studies published between 1990 and 2000 were identified as probably methodically robust and clinically relevant compared with 64.3% for studies published from 2000 to 2004 [29]. The authors concluded that as QoL data and trials improve, they increasingly impact clinical decision making and treatment policies.

Quality of life studies put the patient at the center rather than at the periphery of assessing the effectiveness of treatment interventions. The study of QoL can help to clarify the impact of patient care measures, such as “patient education, pain management techniques, preventive measures for anticipatory nausea with chemotherapy, medications for symptom control, professional counselling about sexual function, family and marital relationships or adjustment to terminal illness” [30]. QoL measures can also provide useful information in issues involving nutritional support and improvements in palliative and home care [31,32]. With this information, it is also possible to investigate some relationships between QoL, treatment regimens and disease outcomes for ovarian cancer patients undergoing chemotherapy. By investigating the important relationships between QoL and chemotherapy treatment, high-risk situations can be better identified and addressed with each individual over the course of their treatment to improve survivorship likelihood for women with ovarian cancer. Also the availability of data documenting the effects and longitudinal changes associated with QoL can contribute to the development of future treatment regimens and approaches to clinical care.

Moreover, QoL data can provide information that can help shape public policy and healthcare decisions made by governmental and private institutions. In addition, it can also guide the research agenda of pharmaceutical companies and cooperative groups.

Quality of life instruments can provide levels of information not always supplied by traditional outcome measures by allowing patients to prioritize which life domains are most important to them. Length of survival alone is an unsatisfactory measure of the success of treatment; the quality of survival needs to be evaluated.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.