Predictive role of vulnerable elderly study-13 screening tools in chemotherapy-related toxicity in elderly patients with lung cancer.

Introduction
Lung cancer is the second most frequently diagnosed cancer in both sexes and the most common cause of cancer-related death. While the risk of lung cancer increases with age, 53% of patients are diagnosed at the age of 70 and over, and 83% are diagnosed at the age of 65 and over1.
Owing to the presence of comorbid diseases, low-performance status, and high psychosocial support needs, cancer treatment is more difficult in elderly patients than in younger patients, and the risk of chemotherapy toxicity is greater2. Despite these high risks, fewer than 25% of patients in cancer-related clinical studies are elderly3. This situation increases the need for tools to guide oncological treatment decisions.
Hurria et al. reported that traditional performance assessments, such as Eastern Cooperative Oncology Group (ECOG) performance status (PS) scores, do not fully predict which elderly patients are at the highest risk for adverse outcomes from chemotherapy4. A study by Williams et al. reported that there were significant differences in the criteria used by oncologists when deciding on chemotherapy for elderly patients and that there was not yet a consensus among physicians on risk factors for this group5. Toxicity prediction tools, in addition to helping clarify the treatment decision-making process, may lead to pretreatment changes, such as reducing the dose or adding a colony-stimulating factor to reduce toxicity6. The comprehensive geriatric assessment (CGA) in patients aged ≥ 65 years receiving chemotherapy is an approved tool for estimating chemotherapy toxicity that conventional assessments conducted by oncologists cannot accurately predict. However, CGA is time-consuming7. In contrast, quick and easy-to-use screening tools that take less than five minutes to administer, such as the Vulnerable Elderly Questionnaire-13 (VES-13) questionnaires, are preferred8,9.
In this observational prospective study, we aimed to investigate the ability of the VES-13 scale to predict treatment-related toxicity in patients with lung cancer receiving chemotherapy. In this way, we want to identify risky patients in clinical applications and determine screening tools that guide treatment practitioners in the treatment of these patients.

Materials and methods
This was a prospective, single-center observational study. All the participating patients provided written consent.
Patients aged 65 years and older who started chemotherapy from newly diagnosed lung cancer patients between December 2022 and December 2023 were included in the study. All patients had at least 2 cycles of adjuvant/neoadjuvant (for Stage 2/3 patients) or metastatic (palliative for Stage 4 patients) treatment regimens planned to be started. Patients with hematological malignancies who received concomitant hormonal therapy, targeted therapy, immunotherapy, or radiotherapy and patients who were mentally unsuitable for consent were excluded from the study.
Eligible patients completed the VES-13 test before starting the first treatment with the help of investigative physicians. Turkish versions of the screening tests were used. The physicians who initiated the treatment regimen were blinded to the test scores.
The primary endpoint of the study was defined as the occurrence of any CTCAE v4.03 grade 3–5 treatment-related toxicity (hematologic or non-hematologic) during chemotherapy and up to 30 days after completion of the planned treatment.

Toxicity assessment and instruments used for assessment

Treatment-related toxicity (TRT)
Patients were followed prospectively from the start of chemotherapy until 30 days after treatment. Patients were questioned about treatment-related toxicity by their attending medical oncologist at the beginning of each treatment cycle and at the end of the planned treatment, and the data were recorded. Clinicopathological patient characteristics, demographic characteristics, laboratory data (laboratory tests were performed at the beginning of treatment and at the beginning of each cycle) and patient follow-up information were recorded. TRTs were rated via the Common Terminology Criteria for Adverse Events (CTCAE) v.4.03. Severe toxicities were considered CTCAE grades 3–5 (requiring discontinuation of treatment, requiring hospitalization, death, etc.). If the patients were exposed to the same toxicity more than once, the highest level of toxicity they experienced was recorded in the study file.
In accordance with the NCCN and EORTC guidelines, the risk of febrile neutropenia was assessed in all patients before their first chemotherapy, and primary prophylaxis was used for patients who had a neutropenic risk of 10% or greater10,11. If G-CSF is not used for primary prophylaxis or when there is neutropenia, it is considered to be ‘use of additional/extra G-CSF.

VES-13
The VES-13 questionnaire consists of 4 groups of questions. Age, self-perceived health, difficulties in performing 6 specific activities, and difficulties in performing daily living tasks due to health concerns from these groups. The score ranges from 0 to 10 points, with ≥ 3 points indicating frailty12. The average time spent is 5–10 min.

Statistical analysis
Descriptive analyses were performed to summarize patient demographics, clinical information, treatment characteristics, and geriatric assessment tests. TRTs were tabulated. Patients were analyzed according to VES-13 (scores less than 3/3 and above) scoring13. The SPSS for Windows 26.0 package program was used for the statistical analysis of the data. Categorical measurements are summarized as numbers and percentages, and continuous measurements are presented as the means and standard deviations. Cross tabulations were used to calculate sensitivity. To evaluate discrimination, a receiver operating characteristic (ROC) curve and the area under the ROC curve (ROC-AUC) were calculated. Logistic regression models were used to compare the survey tools, clinical variables, demographic characteristics, and treatment characteristics between the TRT patients and the questionnaires. Fisher’s exact test and the Mantel‒Haenszel chi-square test were used to evaluate the relationships between interventions for adverse events and VES-13 scores. All statistical tests were two-sided. The statistical significance limit was set at p < 0.05 (α = 0.05), and 95% confidence intervals were calculated.

Results

Patient characteristics
A total of 131 patients were examined. No patient died due to TRT during the follow-up period.
Ninety-nine (75.6%) of the patients were male; the median age of the entire group was 70 (range 65–85) years. Of those included, 67.9% (n = 89) had non-squamous cell carcinoma and 32.1% (n = 42) had squamous cell carcinoma. Twenty-four of the patients (18.3%) were in the nonmetastatic stage (Stage 2: 8(6.1%) and Stage 3: 16(12.2%)), and 87 (66.4%) of the patients had comorbid diseases (such as heart failure, ischemic heart disease, diabetes mellitus, and hypertension) at the beginning of the treatment. Among the patients, 53 (40.5%) received mono-chemotherapy, 78 (59.5%) received poly-chemotherapy, 70 (53.9%) patients started treatment with standard-dose chemotherapy, and 61 (46.6%) patients started modified-dose chemotherapy. Ninety-nine (75.6%) of the patients received first-line chemotherapy. A total of 72 (55%) patients had a BMI below 25, 94 (71.8%) patients had an Eastern Cooperative Oncology Group (ECOG) score below 2, and 62 (47.3%) patients had a VES-13 score below 3.

Treatment-related toxicity (TRT)
The degree of toxicity that developed in a total of 63 (48.1%) patients was recorded as grades 3–5. There was hematologic toxicity in 46 (35.1%) patients and nonhematologic grade 3–5 toxicity in 33 (25.2%) patients. The resulting toxicities and their proportional distributions are shown in Table 1.

Screening results
While the median score in the VES-13 screening was 4, 69 (52.7%) people had impaired VES-13 scores (≥ 3). Grade 3–5 TRT developed in 49 (71%) of those in the impaired VES-13 group. Using a VES-13 cutoff score of ≥ 3, the sensitivity, specificity, positive predictive value, and negative predictive value for predicting grade 3–5 TRT were 77.8%, 70.6%, 71.0%, and 77.4%, respectively. In addition, the ROC-AUC curve was generated. The ROC-AUC value of the VES-13 was 0.715 (95% CI: 0.62–0.81, p < 0.001), which was consistent with its high diagnostic value.
A logistic regression model was established to determine the predictive value of the VES-13 for grade 3–5 TRT. In the univariate regression analysis, VES-13 score ≥ 3 (reference: <3) (OR = 8.40, p < 0.001), ECOG performance status 2–4 (reference: 0–1) (OR = 2.60, p = 0.018), presence of comorbidities (reference: absence of comorbidities) (OR = 2.38, p = 0.024), and use of platinum-based chemotherapy (reference: non-platinum regimens) (OR = 2.53, p = 0.016) were identified as predictive factors for TRT. The results are shown in Table 2. Variables that were significant in the univariate analysis (ECOG performance status, comorbidity status, platinum-based chemotherapy, and VES-13 screening tool) were subsequently included in the multivariate logistic regression model. In the multivariate analysis, VES-13 score ≥ 3 (OR = 8.26, 95% CI: 3.69–18.48, p < 0.001) and platinum-based chemotherapy use (OR = 2.42, 95% CI: 1.03–5.67, p = 0.042) remained independent predictive factors for grade 3–5 TRT.

Owing to toxicity, the treatment of 53 (40.5%) patients was interrupted, the treatment of 27 (20.6%) patients was discontinued, the initial treatment dose was reduced in 32 (24.4%) patients, and extra granulocyte colony stimulating factor (G-CSF) was used in 44 (33.6%) patients. Eighteen (13.7%) patients experienced unexpected hospitalizations. In the analysis, a statistically significant relationship was found between the VES-13 toxicity tool and toxicity interventions (p < 0.05). The results are shown in Table 3.

Discussion
In this study, we investigated the performance of the VES-13 screening tool in predicting TRT in elderly patients with lung cancer receiving chemotherapy. In our study, we found that while severe toxicity occurred in the majority of patients (48.1%), VES-13 could predict severe TRT. Platinum chemotherapy and the Eastern Cooperative Oncology Group (ECOG) performance status along with TRT were also found to be predictive. However, VES-13 remained an independent predictive feature when the predictors were evaluated together.
Studies have reported that geriatric patients are at high risk for toxicity due to their fragile nature and increased sensitivity to toxic exposures and that severe toxicity is observed at least once during treatment14,15. In our study, similar to the literature, grade 3–5 TRT developed in 48.1% of the participants. In a study involving 40 patients with non-small cell lung cancer, the most common hematological toxicity was 80% anemia, and the most common nonhematological toxicity was nausea and vomiting16. In our study, the most common hematological toxicity was neutropenia (22.8%), and the most common nonhematological toxicity was fatigue (7.6%). We believe that this difference in TRT may be due to the regimen used due to the inclusion of a single cancer type in our study, ethnic and geographical differences affecting the pharmacokinetics of chemotherapy drugs, and differences in regional dietary habits17–19.
In the study of Kumar et al., one of the previous studies in the literature, VES-13 was found to be associated with both TRT and TRT-related death20. In the studies of Luciani et al. and Ferrero et al. conducted in Italy, VES-13 was reported to be predictive of TRT in cancer patients21,22. In a study conducted in our country that included 208 nonhematological cancer patients, VES-13 was found to be a strong predictor of chemotherapy toxicity13. The VES-13 was not originally a scale for cancer patients. However, studies conducted in recent years have revealed that this scale is important for cancer patients. VES-13 is recommended for toxicity in the current National Comprehensive Cancer Network (NCCN) guidelines (ASCO) and, finally, in the advanced lung cancer consensus in China in 202223,24. The results of our study also showed that VES-13 score is a strong predictor of TRT.
In our study, a significant relationship was found between medical interventions given to patients exposed to toxicity and VES-13. Similar to our study, in a study conducted in Turkey with a large number of patients, VES-13 was associated with unexpected hospitalization, treatment modification and additional G-CSF13. Another study involving patients with gynecological malignancies revealed a significant association of VES-13 in the group requiring additional G-CSF use with the discontinuation of treatment22. These results are acceptable since VES-13 predicts toxicity. Importantly, unlike previous studies, the toxicity scale in this study also predicted blood transfusion.

Conclusion
In conclusion, we found that the VES-13 toxicity screening tool can be a useful tool for making chemotherapy decisions, choosing the appropriate regimen, and predicting treatment-related adverse events during chemotherapy in geriatric patients with lung cancer in Türkiye. We believe that the use of the VES-13 questionnaire in daily oncology practice is beneficial for clinicians. Importantly, this study provides prospective, tumor-specific evidence demonstrating that VES-13 independently predicts not only severe chemotherapy-related toxicity but also clinically meaningful toxicity-driven interventions, such as treatment modification, hospitalization, and supportive care requirements, in a real-world geriatric oncology population.

Limitations
The prospective design of our study, which included a single cancer group and an advanced-age patient group, is its strengths. However, there are several limitations in our study. The fact that this was a single-center study is a limiting factor in the analysis that screening tools assess only the predictive potential for Grade 3–5 toxicity, which is characterized as severe adverse events. For the geriatric population, lower-grade toxicities may also have important consequences.