1. Introduction
Predicting chemotherapy-related toxicities is a key benefit of integrating geriatric assessments (GA) in the routine care of older adults with cancer. Some toxicity prediction models have been developed to better inform chemotherapy decisions in this setting and guide discussions with older patients. Table 1 summarizes the features of three available chemotherapy toxicity prediction tools for older adults with cancer to predict severe chemotherapy toxicity.
Table 1.
Key characteristics of the available chemotherapy toxicity prediction tools.
| Characteristics | CRASH | CARG | CARG-BC | |
|---|---|---|---|---|
|
| ||||
| Study inclusion criteria | Age ≥ 70 years Histologically proven cancer Scheduled to receive a new chemotherapy regimen in 1st-4th line No diagnosed with dementia Not requiring concomitant radiotherapy |
Age ≥ 65 years Diagnosis of cancer Scheduled to receive a new chemotherapy regimen |
Age ≥ 65 years Diagnosis of stage I-III breast cancer Scheduled to receive an adjuvant or neoadjuvant chemotherapy |
|
| Key study derivation cohort characteristics | Number of patients | 331 | 500 | 283 |
| Age (range), years | Median: 76 (70–92) | Mean: 73.1 (65–91) | Median: 70.(65–85) | |
| Sex | • Women: 50.2% • Men: 49.8% |
• Women: 56% • Men: 44% |
• Women: 98.9% • Men: 1.1% |
|
| Cancer diagnoses | •Lung: 21.5% •Breast: 21.5% •Non-Hodgkin lymphoma: 14.2% •Colorectal: 12.4% •Bladder: 6.9% •Other: 24.4% |
•Lung: 29% •Breast: 11% •Gastrointestinal: 27% •Gynecological: 17% •Genitourinary: 10% •Other: 6% |
• Breast: 100% ○ Triple-negative: 24.4% ○ HER2-negative HR-positive: 48.4% ○ HER2-positive/HR-positive: 17.0% ○ HER2-positive/HR-negative: 10.2% |
|
| Cancer stage | • I: 6.3% • II: 14.5% • III: 23.6% • IV: 55.6% |
• I: 5% • II: 12% • III: 22% • IV: 61% |
• I: 38.9% • II: 40.6% • III: 20.5% |
|
| Treatment intent | • Curative: 45.3% • Palliative: 54.7% |
Not available | • Curative: 100% ○ Neoadjuvant: 17.7% ○ Adjuvant: 82.3% |
|
| Key study validation cohort characteristics | Number of patients | 187 | 250 | 190 |
| Age (range), years | Median: 75 (70–89) | Mean: 73.0 (65–91) | Median: 70 (65–86) | |
| Sex | • Women: 50.8% • Men: 49.2% |
• Women: 55% • Men: 45% |
• Women: 100% • Men: 0% |
|
| Cancer diagnoses | • Lung: 19.8% • Breast: 17.6% • Non-Hodgkin lymphoma: 16.6% • Colorectal: 10.2% • Bladder: 7.0% • Other: 28.8% |
• Lung: 26% • Breast: 24% • Gastrointestinal: 27% • Gynecological: 7% • Genitourinary: 12% • Other: 4% |
• Breast: 100% ○ Triple-negative: 22.6% ○ HER2-negative HR-positive: 50.0% ○ HER2-positive/HR-positive: 20.0% ○ HER2-positive/HR-negative: 8.4% |
|
| Cancer stage | • I: 3.2% • II: 18.2% • III: 25.1% • IV: 53.5% |
• I: 4% • II: 16% • III: 27% • IV: 52% |
• I: 32.1% • II: 46.4% • III: 21.6% |
|
| Treatment intent | • Curative: 44.9% • Palliative: 55.1% |
Not available | • Curative: 100% ○ Neoadjuvant: 16.8% ○ Adjuvant: 83.2% |
|
| Items | Geriatric assessments | IADL: • <26 = 1 • ≥26 = 0 ECOG PS: • 1–2 = 1 • 3–4 = 2 Mini Mental State Examination: • <30 = 2 • ≥30 = 0 Mini Nutritional Assessment: • <28 = 2 • ≥28 = 0 |
Hearing: • Excellent/good = 0 • Fair/poor/totally deaf = 2 Falls during last 6 months: • 0=0 • ≥1=3 IADL: taking medications: • Without help = 0 • With some help/completely unable = 1 Walking limitations one block: • Not limited at all = 0 • Limited a little/a lot = 2 Decreased social activity: • None/a little of the time = 0 • Some/most/all of the time = 1 |
Falls during last 6 months: • 0 = 0 • ≥1=4 Walking limitations ≥1 mile: • Not limited at all = 0 • Somewhat/very limited = 3 Availability of someone to provide advice about a crisis: • Most/all of the time = 0 • None/little/some of the time = 3 |
| Patient characteristics | – | Age: • <72 years = 0 • ≥72 years = 2 |
– | |
| Clinical parameters | Diastolic blood pressure: >72 mmHg = 1 | – | – | |
| Laboratory parameters | LDH: >459 U/L = 1 | Hemoglobin: • ≥11 g/dL (male) or ≥ 10 g/dL (female) = 0 • <11 g/dL (male) or < 10 g/dL (female) = 3 Creatinine clearance (Jeliffe formula): • ≥34 mL/min = 0 • <34 mL/min = 3 |
Hemoglobin: • >13 g/dL (male) or > 12 g/dL (female) = 0 • ≤13 g/dL (male) or ≤ 12 g/dL (female) = 3 Liver function tests: • Within normal reference range = 0 • Outside normal reference range = 3 |
|
| Disease characteristics | – | Cancer type: • Other = 0 • Gastrointestinal or genitourinary = 2 |
Stage: • I = 0 • II-III = 2 |
|
| Treatment features | Chemotoxicity | Chemotherapy dose: • Reduced = 0 • Standard = 2 Number of chemotherapy agents: • Mono-chemotherapy = 0 • Polychemotherapy = 2 |
Anthracycline use: • No = 0 • Yes = 1 Planned chemotherapy duration • ≤3 months = 0 • >3 months = 4 |
|
| Outcomes | Grade 4 hematological toxicity • Grade 3–4 non-hematological toxicity Combined toxicity |
Grade 3–5 toxicity | Grade 3–5 toxicity | |
| Proportion of patients experiencing toxicities for each risk category | Category | Heme | Non-heme | Combined | Total risk score | Risk | Total risk score | Risk | ||||
| Score | Risk | Score | Risk | Score | Risk | Low | 0–3 | 25% | 0–5 (low) | 22% | ||
| Low | 0–1 | 7% | 0–2 | 33% | 0–3 | 50% | 4–5 | 32% | ||||
| Intermediate-low | 2–3 | 23% | 3–4 | 46% | 4–6 | 58% | Medium | 6–7 | 50% | 6–11 | 51% | |
| Intermediate-low | 4–5 | 54% | 5–6 | 67% | 7–9 | 77% | 8–9 | 54% | (intermediate) | |||
| High | >5 | 100% | >6 | 93% | >9 | 79% | High | 10–11 12–19 |
77% 89% |
≥12 (high) | 81% | |
CRASH = Chemotherapy Risk Assessment Scale for High Age; CARG = Cancer and Aging Research Group; CARG-BC = Cancer and Aging Research Group – Breast Cancer; IADL = Instrumental activities of daily living; ECOG PS = Eastern Cooperative Oncology Group Performance Status, LDH = lactate dehydrogenase.
1.1. Chemotherapy Risk Assessment Scale for High Age score
The Chemotherapy Risk Assessment Scale for High Age (CRASH) score was developed by investigators at the Senior Adult Oncology Program of the Moffitt Cancer Center to predict the risk of severe chemotherapy-related hematologic and non-hematologic toxicity in older individuals with cancer [1]. This prediction model includes the specific chemotherapy regimen being considered, along with laboratory values (creatinine, albumin, hemoglobin, lactate dehydrogenase [LDH], liver function tests) and geriatric assessments variables (function, cognition, and nutrition) (Table 1). The score was developed and validated in a cohort of 518 patients aged ≥70 including 54.8% of individuals with advanced malignancies, including 20.8% with lung cancer, 20.1% with breast cancer, 15.1% with non-Hodgkin lymphoma, and 11.6% with colorectal cancer.
In this cohort, predictors of grade 4 hematologic toxicity included lymphocyte count, aspartate aminotransferase level, Instrumental Activities of Daily Living (IADL) score, LDH level, diastolic blood pressure, and toxicity of the chemotherapy regimen. Predictors of grade 3–4 non-hematologic adverse events included hemoglobin, creatinine clearance, albumin, self-rated health, Eastern Cooperative Oncology Group (ECOG) Performance Status (PS), Mini-Mental Status score, Mini-Nutritional Assessment score, and toxicity of the chemotherapy regimen. The combined hematologic and non-hematologic risk categories identified patients with low, medium-low, medium-high, and high risk experiencing adverse events in 50%, 58%, 77%, and 79% of cases, respectively. Nonetheless, the split score discriminated better compared with the combined score.
1.2. Cancer and Aging Research Group (CARG) score
The Cancer and Aging Research Group (CARG) score was developed by researchers in the United States to predict the risk of experiencing severe or fatal toxicity due to chemotherapy [2], defined as grade 3–5 based on the Common Terminology Criteria for Adverse Events (CTCAE). This model considers age, type of malignancy, the chemotherapy regimen under consideration, hemoglobin, creatinine clearance, hearing, ability to take medications, physical activity, and social activity (Table 1). The score was validated in a cohort of 250 patients aged ≥65 years with solid tumors, mostly including breast cancer (24%), lung cancer (26%), and gastrointestinal malignancies (27%) [3].
In the overall population, 58% of patients experienced grade ≥ 3 adverse events in the validation study. The CARG score was associated with the risk of severe toxicity, which occurred in 37% of patients with a low score versus 62% of those with medium score, and 70% of those with high score. Of note, this model significantly outperformed clinician-rated Karnofsky Performance Status (KPS) in predicting adverse events associated with chemotherapy in its independent validation study.
1.3. Cancer and Aging Research Group-Breast Cancer score
Recently, the CARG-Breast Cancer (CARG-BC) score was modelled to predict severe and fatal complications related to chemotherapy use in older patients with curable breast cancer [4]. Similar to the CARG score, this model includes breast cancer stage, chemotherapy regimen and its duration, along with liver function tests, hemoglobin, falls, limited mobility, and social support (Table 1). The score was developed and validated in a cohort of 473 patients with stage I-III breast cancer, including 23.7% with triple-negative disease, and 27.7% with human epidermal growth factor receptor 2 (HER2) cancer.
In the study validation cohort, the rates of CTCAE grade 3–5 toxicity were 27%, 45% and 76% in patients with low, medium, and high-risk score, respectively. These categories also correlated with hospitalization rate and reduced chemotherapy dose intensity.
The case example in Table 2 illustrates how all three scores can predict severe chemotherapy toxicity.
Table 2.
Case example comparing the use of the three chemotherapy toxicity prediction tools.
| Characteristics | CRASH | CARG | CARG-BC | |
|---|---|---|---|---|
|
| ||||
| Items | Geriatric assessments | IADL: • <26 = 1 ECOG PS: • 1–2 = 1 Mini Mental State Examination: • <30 = 2 Mini Nutritional Assessment: • <28 = 2 |
Hearing: • Excellent/good = 0 Falls during last 6 months: • 0 = 0 IADL: taking medications: • With some help/completely unable = 1 Walking limitations one block: • Limited a little/a lot = 2 Decreased social activity:• Some/most/all of the time = 1 |
Falls during last 6 months: • 0 = 0 Walking limitations ≥1 mile: • Somewhat/very limited = 3 Availability of someone to provide advice about a crisis: • Most/all of the time = 0 |
| Patient characteristics | – | Age: • <72 years = 0 |
– | |
| Clinical parameters | Diastolic blood pressure: >72 mmHg = 1 | – | – | |
| Laboratory parameters | LDH: <459 U/L = 0 | Hemoglobin: • ≥11 g/dL (male) or > 10 g/dL (female) = 0 Creatinine clearance (Jeliffe formula): • ≥34 mL/min = 0 |
Hemoglobin: • >13 g/dL (male) or > 12 g/dL (female) = 0 Liver function tests: • Within normal reference range = 0 |
|
| Disease characteristics | – | Cancer type: • Other = 0 |
Stage: • II-III = 2 |
|
| Treatment features | Chemotoxicity | Chemotherapy dose: • Standard = 2 Number of chemotherapy agents: • Polychemotherapy = 2 |
Anthracycline use: • Yes = 1 Planned chemotherapy duration • >3 months = 4 |
|
| Score ➔ Predicted Outcomes | Heme Score: 2 ➔ Grade 4 hematological toxicity = 23% Non-heme Score: 5 ➔ Grade 3–4 non-hematological toxicity = 67% Combined Score: 7 ➔ Combined toxicity = 77% |
Score: 8 ➔ Grade 3–5 toxicity = 54% | Score: 10 ➔ Grade 3–5 toxicity = 51% | |
Seventy year old white female with pT3 N1, triple receptor-negative breast cancer, who had a recent mastectomy and axillary node clearance. Doublet-agent sequential chemotherapy including a standard dose of an anthracycline and a taxane is being considered for 6 months with curative intent. She requires some help with traveling to places, shopping, preparing meals, doing housework, laundry, taking medications, and managing money. She can use the telephone independently. On Mini-Mental State Exam, she loses 2 points for recall and 2 points in attention and calculation. Her ECOG performance status is 1. Mini nutritional assessment score is 27. She has good hearing, has had no falls, but limited a little for walking a block. She has some decreased social activity but has excellent social support from her children. Diastolic blood pressure is 80 mmHg. Labs: LDH normal, hemoglobin 13 g/dL, Creatinine Clearance 55 mL/min, liver function tests normal.
CRASH = Chemotherapy Risk Assessment Scale for High Age; CARG = Cancer and Aging Research Group; CARG-BC = Cancer and Aging Research Group – Breast Cancer; IADL = Instrumental activities of daily living; ECOG PS = Eastern Cooperative Oncology Group Performance Status, LDH = lactate dehydrogenase.
Comment: These are prediction tools and used to guide a discussion. The provider should use the tool that is feasible in their practice and counsel the patient accordingly. These tools should not drive treatment decisions, but rather inform discussions with patients (as they do not include patient preferences); 2) clinicians should use the tool that they feel comfortable with or that is feasible in their own practice and has been validated in a population of patients similar to those that they are seeing; and 3) there are no prospective data comparing their performance, which is an area of needed research.
2. Independent validation studies
Efforts have been made to validate these chemotherapy prediction tools and support their role to inform shared decision-making in more diverse populations recruited in different geographical areas. Importantly, the CARG and the CRASH tool have been compared in an analysis of 120 German patients with mean age of 77.2 years and mostly diagnosed with solid tumors (57%) [5]. In this cohort, the predictive performance of the CARG and the combined CRASH was similar for overall and hematologic toxicity.
No correlation of the CARG score categorization with grade ≥ 3 toxicity was confirmed also in a prospective study recruiting 126 patients aged ≥65 years receiving chemotherapy in two Australian hospitals [6]. These patients were diagnosed mostly with colorectal cancer (36%), upper gastrointestinal cancer (18%), and thoracic malignancies (10%), and suitable for palliative systemic therapy in 65% of cases.
In another study of 64 men with metastatic castration-resistant prostate cancer receiving chemotherapy in 3 academic centers in Canada (mean age of 73 years and most received docetaxel), The predictive role of the CARG score in this population (low risk: 22%; moderate risk: 53%; high risk: 71%) was confirmed [7]. However, the investigators did not document any association of the score with grade 2 adverse events, that may be important to older adults with cancer. The CRASH score was included in the PANDA study (NCT02904031) that compared FOLFOX plus panitumumab to 5-fluorouracil plus panitumumab, which is recruiting older patients with RAS and BRAF wild-type advanced colorectal cancer aged ≥70 years [8].
3. Additional evidence
A number of additional studies have investigated the performance of the available chemotherapy toxicity prediction tools. However, they have included smaller cohorts of patients, provide limited details and/or are published as meeting abstracts. For example, some studies have failed to confirm the predictive role of these tools. The CRASH and the CARG scores have been retrospectively evaluated in 248 patients aged ≥70 years recruited in the Elderly Cancer Patients (ELCAPA) prospective study [9]. Nonetheless, the ELCAPA study allowed changes in the treatment plan based on a geriatric-led CGA and this occurred in 78 (21%) of patients recruited in 2007–2009. The CARG did not show an association of CARG score with severe chemotherapy-related toxicity in 259 patients aged ≥65 years diagnosed with solid tumors and receiving chemotherapy in a single institution in Hong Kong [10].
Conversely, some of these studies have confirmed the validity of these tools. A retrospective analysis included 599 Italian patients treated in two Institutions with a median age of 76.4 years mostly diagnosed with lung cancer (37%) and colorectal cancer (26%)and documented a significant association of the CARG score with hematologic and non-hematologic toxicity [11]. The utility of the CRASH and CARG score was also documented in an Italian cohort of 62 patients with advanced breast cancer treated at a single institution [12].
Some of these analyses have also evaluated the validity of the CARG tool in a diverse range of geographical areas [13]. On the other hand, some analyses have assessed their role in patients receiving different anti-cancer treatments, such as a study of 85 patients with locally advanced non-small-cell lung cancer undergoing chemoradiotherapy, for whom the CARG categories did not correlated with grade 3–4 toxicity [14].
Although the quality of this evidence is more limited, these analyses suggest that there is an increasing global interest on chemotherapy toxicity prediction tools to better inform discussions with older patients with cancer.
4. How to implement chemotherapy toxicity tools into clinical practice
These tools should be performed at the initial or subsequent oncology visits, when discussions about chemotherapy occur. Also, these tools can aide in clinical decision-making regarding single versus doublet chemotherapy and whether to start with a dose reduction. Among these tools, there may be some variability in prediction scores, as outlined in the case in Table 2. This may pose a challenge in interpretation. If a score shows a high risk, it should not be interpreted as a deterrent to chemotherapy. Vice versa, if there is a low risk, it should not be interpreted as a clearance for chemotherapy. Therefore, these tools should be used to better inform and enhance discussions with older patients with cancer around chemotherapy decisions as their preferences should remain the key driver. While some individuals may perceive the predicted risk of toxicity on chemotherapy not worth the potential benefits, other patients may perceive the balance of potential risks and benefits differently. These tools are for guidance and older adults pursuing chemotherapy should still be monitored closely and proactively for toxicity. Clinicians should use the tool that they feel comfortable with or that is feasible in their own practice and has been validated in a population of patients similar to those that they are seeing; and 3) there are no prospective data comparing their performance, which is an area of needed research.
5. Limitations and gaps of knowledge on the current chemotherapy toxicity prediction tools
As illustrated above, chemotherapy toxicity tools are helpful to predict severe toxicity from chemotherapy in older adults with cancer, and therefore, inform decision making for chemotherapy intensification as well as early interventions and closer monitoring during treatment. However, there are some gaps in the current literature that warrant further research.
Given the heterogeneity of populations and their specific needs, chemotherapy toxicity tools may not be applicable to different cancer patient populations. As we adopt the tools above into clinical practice, we must also confirm the validity in various populations across the world. Depending on the special needs of a community (i.e., resources in developed versus developing countries), tools specific to a community may need to be developed and validated.
In view of the unique needs of each cancer type, the development of chemotherapy toxicity tools specifically tailored towards disease types are necessary, especially in disease types where there is an overlap in symptoms from the advanced disease and geriatric syndromes (i.e., cancer cachexia and nutritional status in GI cancers). Further, these toxicity tools are limited to patients receiving chemotherapy. With the advancement of drug development beyond chemotherapy to targeted therapies and immunotherapy, tools predicting specific toxicities to these agents should be developed and studied in older adults with cancer. We need more studies, like the PANDA study [8], that incorporate these tools into prospective cancer treatment clinical trials.
Additionally, in real world practice, balancing the risk/benefit of toxicity in the curative setting for early-stage disease can pose a key challenge than in the palliative care setting. Tools like the CARG-BC in early-stage breast cancer can help to assist with decision-making specific to this population. This newer chemotherapy toxicity tool serves as an example of how we should explore similar tools in other early-stage cancers.
Further, given current time and resource constraints, it may be challenging to optimize resources to utilize chemotherapy toxicity tools in daily practice. Whether we should use these tools routinely for every patient or only when we are considering a specific patient population (i.e., focusing on vulnerable versus all older adults) can be personalized to each practice depending on the availability of resources. In some settings, training other members of the health care team beyond the physician can be essential for incorporating these tools into a clinical practice.
6. Conclusion
For prediction of severe chemotherapy toxicity, these tools are important resources to incorporate into the care of older adults with cancer. Given drug development has moved beyond chemotherapy with newer agents, the scope of the current tools is limited to those receiving chemotherapy at this time. Therefore, further research is necessary to predict toxicities from newer agents, and therefore, develop interventions to prevent or lessen toxicities.
Footnotes
Declaration of competing interest
No conflicts of interest or disclosures.
References
- [1].Extermann M, Boler I, Reich RR, Lyman GH, Brown RH, DeFelice J, et al. Predicting the risk of chemotherapy toxicity in older patients: the Chemotherapy Risk Assessment Scale for High-Age Patients (CRASH) score. Cancer. 2012;118(13):3377–86. [DOI] [PubMed] [Google Scholar]
- [2].Hurria A, Togawa K, Mohile SG, Owusu C, Klepin HD, Gross CP, et al. Predicting chemotherapy toxicity in older adults with cancer: a prospective multicenter study. J Clin Oncol. 2011;29(25):3457–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [3].Hurria A, Mohile S, Gajra A, Klepin H, Muss H, Chapman A, et al. Validation of a prediction tool for chemotherapy toxicity in older adults with cancer. J Clin Oncol. 2016; 34(20):2366–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Magnuson A, Sedrak MS, Gross CP, Tew WP, Klepin HD, Wildes TM, et al. Development and validation of a risk tool for predicting severe toxicity in older adults receiving chemotherapy for early-stage breast Cancer. J Clin Oncol. 2021;39(6):608–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].Ortland I, Mendel Ott M, Kowar M, Sippel C, Jaehde U, Jacobs AH, et al. Comparing the performance of the CARG and the CRASH score for predicting toxicity in older patients with cancer. J Geriatr Oncol. 2020;11(6):997–1005. [DOI] [PubMed] [Google Scholar]
- [6].Moth EB, Kiely BE, Stefanic N, Naganathan V, Martin A, Grimison P, et al. Predicting chemotherapy toxicity in older adults: comparing the predictive value of the CARG toxicity score with oncologists’ estimates of toxicity based on clinical judgement. J Geriatr Oncol. 2019;10(2):202–9. [DOI] [PubMed] [Google Scholar]
- [7].Alibhai SMH, Breunis H, Hansen AR, Gregg R, Warde P, Timilshina N, et al. Examining the ability of the Cancer and Aging Research Group tool to predict toxicity in older men receiving chemotherapy or androgen-receptor-targeted therapy for metastatic castration-resistant prostate cancer. Cancer. 2021;127(14):2587–94. [DOI] [PubMed] [Google Scholar]
- [8].Battaglin F, Schirripa M, Buggin F, Pietrantonio F, Morano F, Boscolo G, et al. The PANDA study: a randomized phase II study of first-line FOLFOX plus panitumumab versus 5FU plus panitumumab in RAS and BRAF wild-type elderly metastatic colorectal cancer patients. BMC Cancer. 2018;18(1):98. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Frelaut M, Caillet P, Culine S, Paillaud E, Tournigand C, Canoui-Poitrine F. External validation of two predictive scores of chemotherapy toxicities among older patients with solid cancer, from ELCAPA prospective cohort. J Clin Oncol. 2020;38(15_suppl): 12050. [Google Scholar]
- [10].Chan WL, Ma T, Cheung KL, Choi H, Wong J, Lam KO, et al. The predictive value of G8 and the Cancer and aging research group chemotherapy toxicity tool in treatment-related toxicity in older Chinese patients with cancer. J Geriatr Oncol. 2021;12(4): 557–62. [DOI] [PubMed] [Google Scholar]
- [11].Luciani A, Brunello A, Battisti NML, Romanato G, Caldiera S, Bergamo F, et al. The assessment of chemotherapy risk in elderly cancer patients: validation of the CRASH score in an Italian cohort. J Clin Oncol. 2015.;33(15_suppl) e20521-e. [Google Scholar]
- [12].Carreca IU, Bazan V, Galvano A, Carreca AP, Cusenza S, Russo A. Advanced breast cancer at presentation (ABC-p) in octogenarian women (OW): specific elderly-devoted risk tests(SEDRT) (CARG+CRASH) as new tools to prevent serious/irreversible adverse reactions (AR) in frail patients. J Clin Oncol. 2021.;39(15_suppl) e12510-e. [Google Scholar]
- [13].Gattani S, Noronha V, Ramaswamy A, Castelino R, Nair V, Nookala M, et al. Accuracy of the CARG chemotherapy toxicity risk prediction tool in older Indian patients with cancer. J Clin Oncol. 2021.;39(15_suppl) e24023-e. [Google Scholar]
- [14].Nadal E, Ruffinelli JC, Palmero R, Navarro A, Plana M, Barroso CM, et al. Geriatric assessment to predict toxicity in elderly patients with unresectable locally advanced non-small-cell lung cancer treated with concurrent chemoradiotherapy. J Clin Oncol. 2018;36(15_suppl):8557. [Google Scholar]