Our findings show that the G8 test score has high power in detecting CGA impairment and probable sarcopenia. Moreover, this tool is successful in identifying probable sarcopenia exclusively. To the best of our knowledge, this study is the first to evaluate probable sarcopenia using the G8 screening exclusively and in conjunction with CGA.
Preparing treatment plan for cancer patients requires careful consideration of several factors, such as the patients’ physical status, performance, cognitive function, nutritional status, frailty, sarcopenia, and risk of chemotherapy-induced toxicities. All these factors directly or indirectly affect the mortality and morbidity of the patient[16, 18, 30]. CGA and sarcopenia detection are helpful tools for physicians to identify patients who require additional care. However, CGA is a time-consuming approach and requires specially trained staff to conduct it.[17]
Ryan et al. showed that sarcopenia is associated with reduced tolerance to chemotherapy, increased postoperative complications, impaired quality of life, and shorter lifespan in cancer patients[31]. Of the different methods used to detect sarcopenia, tests, such as Dual-Energy X-ray Absorptiometry (DEXA), Computed Tomography (CT), Magnetic Resonance Imaging (MRI) that show muscle mass, are considered the gold standard. In the absence of these diagnostic tools, handgrip strength can be used to detect probable sarcopenia[28]. Kilgour et al. showed that handgrip strength was independently associated with survival, functionality, and quality of life[32]. Our results corroborate these findings as we detected that patients with muscle strength also had a low G8 score. Moreover, we showed that as the G8 score decreased, the number of falls in a year increased, whereas the BADL, IADL, and cognition scores reduced. Therefore, we infer that the G8 test score is associated with patients' quality of life and functional independence. Considering the screening power of G8 score for sarcopenia when the cut-off value was taken as 12.5, the sensitivity was low (50%) but the specificity was high (92%). This could lead to a low number of false positives but a large number of false negatives because of the high specificity.
We found a strong correlation between the G8 test and the CGA (κκ 0.638; p < 0.001). To consider the G8 test as reliable, it should provide consistent results when performed by different persons and by the same person at different times. Both the inter-clinician assessments and the assessments repeated by the same clinician at different times provided consistent results and were strongly correlated with the CGA. While investigating the validity and reliability of the G8 geriatric screening test in cancer patients, the calculated cut-off values should be compared with those referred in the original study. Although the cut-off value in our study (≤ 14) according to the ROC analysis was similar to that in the original study[7], the specificity (95%) and sensitivity (79%) rates found in our study slightly differed from those seen in previous studies. In the study conducted by Bellera et al.[7] with the same cut-off value (14), the sensitivity and specificity rates were 85% and 65%, respectively. Although the sensitivity rates were comparable, the specificity rate was higher in our study. During the calculation of abnormal CGA, Bellera et al.[7] included the deterioration in Cumulative Illness Rating Scale-Geriatric (CIRS-G) and timed up and go tests into the calculation in addition to the test used by us. The differences between the sensitivity and specificity rates in these two studies may depend on this factor. The cut-off value was 12.5 in the study conducted by Baitar et al., and the components of GCA, as well as the definition of abnormal CGA used by them differed from ours; they used the GDS-30, social support index, and timed up and go tests for CGA.
The fact that the G8 test detects those with both probable sarcopenia and impaired CGA (AUC = 0.939) indicates that the detection rate of the test is outstanding. Although its sensitivity and specificity were found to be adequate (93%, 86%, respectively), its negative predictive value of 98.1%, which detects the distinction of healthy individuals, could qualify the G8 screening tool as a potentially ideal screening test[33].
In our study, we also investigated the correlation of the G8 test scores with the sub-groups of the CGA. There was a moderate-to-high correlation between the G8 test and MNT-SF. Some additions were made to the MNT-SF test during the original development of the G8 test, and a significant correlation was then expected[7]. Malnutrition screening is critical because existing malnutrition may worsen because of the cancer[34].
The BADL and IADL are important for determining the dependency of the patient and may be useful for deciding the cancer treatment, predicting chemotherapy-related side effects and determining the life expectancy[35]. We found a weak-to-moderate correlation between BADL and G8 geriatric screening test and a moderate correlation between IADL and the G8 test. Studies have demonstrated that the functional impairments were related to the shorter life expectancy[35–37].
The rate of cognitive dysfunction changes between 24% and 38% in older adult cancer patients. The cognitive level of the cancer patient may play an important role in deciding the treatment plan, evaluating chemotherapy-induced toxicity, and determining the most appropriate approach together with the patient[38]. In addition, the survival and mortality rates are significantly affected by impaired cognitive function and low MMSE scores[39, 40]. Therefore, cognitive functions should definitely be assessed in cancer patients[38]. MMSE is an ideal tool for assessing cognitive functions in cancer patients, and our study showed that the G8 geriatric screening test had a moderate consistency with the MMSE results.[22, 38]
Furthermore, the correlation between the G8 test and GDS-15 is an important factor to be taken into consideration. Cancer leads to many adverse changes in the patient’s life. Limitations in social life, job loss, sexual dysfunction, change in the meaning of life, and financial problems may arise and negatively affect the psychological status of the patient[41]. The concomitant depression in older adults with cancer worsens the loss of appetite, weight loss, and malnutrition associated with malignancy. Mood disorders decrease the functionality and quality of life and more importantly, compliance with the cancer treatment[42]. In older adults, depression is also associated with increased mortality[41, 43–45]. Weinberger et al. emphasize that depression was very common among cancer patients and led to negative outcomes[46]. The American Society of Clinical Oncology recommended a routine and regular examination of the mood status in cancer patients[41]. In light of this recommendation, the correlation that we found between the G8 geriatric screening test and GDS-15 can be regarded an important finding.
Another important aspect of this study is its novelty in evaluating the usability of the G8 screening test in detecting probable sarcopenia. Although the specificity of the G8 screening test in detecting probable sarcopenia is high, the high sensitivity and specificity of this test in evaluating CGA and sarcopenia together is particularly notable. The study’s inadequacies comprise the small sample size and evaluation of only probable sarcopenia. Besides, the concomitant co-morbidities could be included in CGA with the scores calculated with the help of one of the co-morbidity indexes. Future studies may compare the G8 screening test with the Turkish validated SARC-f screening test to further measure the formers’ screening power. It may also be useful to consider the performance of the G8 test in the future to evaluate sarcopenia by using gold standard measurement methods, such as CT, MRI, and DEXA.
Our secondary aim was to determine the validity and reliability of the G8 test. Our results show that compared with the CGA, the Turkish version of the G8 test is a valid and reliable screening tool for solid cancer patients. The inclusion of different types of solid cancers in this study improves the generalizability of our findings. Further, the implementation of CGA by a geriatrist instead of specialists from other fields increases the reliability of the tests used. The correlations of the G8 test with both the CGA and its individual components determined in our study, indicate that this test can be safely used in cancer patients in our country.
One of the limitations of our study is the small sample size. Second, we did not use gold standart methods to detect sarcopenia. Also, performance tests could be evaluated in the patients to define severity of the sarcopenia.