The effect of 5-item modified frailty index on clinical outcomes in elderly rectal cancer patients after curative surgery

doi:10.21203/rs.3.rs-4866682/v1

Frailty is a significant factor contributing to an increased risk of adverse clinical outcomes in older cancer patients. This study aims to investigate the impact of the 5-item modified frailty index (mFI) on both short-term and long-term prognosis in elderly rectal cancer patients who underwent radical resection. And, by integrating the 5-item mFI with clinicopathological characteristics, a postoperative major complications nomogram and a cancer-specific survival (CSS) prediction nomogram were further constructed. Demographic, clinical, and therapeutic data were collected from 1,034 patients aged ≥ 70 years with rectal cancer who underwent proctectomy between 2017 and 2022. Patients were categorized into three groups based on their mFI scores: 318 patients with mFI = 0, 404 patients with mFI = 1, and 312 patients with mFI ≥ 2. Comparisons among these groups revealed that higher 5-item mFI scores were associated with an increased incidence of both overall and major postoperative morbidity, prolonged recovery times, and elevated total medical costs. Multivariate logistic regression analysis indicated that an mFI score of ≥ 2 [odds ratio = 3.145, 95% confidence interval (CI): 2.047–4.833, P < 0.001] was an independent risk factor for major postoperative complications. Similarly, in competing risk analysis, the 5-item mFI was identified as an independent prognostic factor for CSS (subdistribution hazard ratio = 2.00, 95% CI: 1.47–2.72, P < 0.001). The postoperative major complications nomogram and CSS prediction nomogram AUC values were 0.730 and 0.844, respectively, both demonstrating superior predictive capabilities. In conclusion, the 5-item mFI serves as a concise and effective tool for preoperative risk stratification and for predicting clinical outcomes in elderly rectal cancer patients.

Biological sciences/Cancer/Gastrointestinal cancer/Colorectal cancer/Rectal cancer

Health sciences/Oncology/Surgical oncology

In recent years, the increasing prevalence of colorectal cancer (CRC) has emerged as a significant challenge for public health, particularly in the context of an aging population¹. However, the risk assessment, diagnosis, and treatment of CRC in elderly patients have not received adequate attention. The clinical outcomes for these patients are highly variable due to the presence of complex underlying diseases, numerous comorbidities, and suboptimal nutritional status. Most prospective randomized controlled trials tend to impose strict restrictions on the physiological reserve of participants to mitigate heterogeneity and bias in the results. Consequently, there is a notable lack of high-quality, evidence-based medical research specifically addressing the needs of elderly patients with CRC.

Frailty, recognized as a geriatric medical syndrome, is characterized by the accumulation of multiple functional impairments and declined multiple interrelated physiological system, which is associated with adverse clinical outcomes^2–4. Findings of previous studies demonstrated that preoperative debilitating condition is significantly associated with post-surgical morbidity and mortality and poor long-term prognosis^5–7. However, there is no uniform consensus on a 'gold standard' for preoperative frailty assessment tools in elderly cancer patients. The frailty index, the representative of the cumulative model of frailty, was first proposed by Rockwood et al. in the Canadian Study of Health and Aging⁸. Despite its utility, the frailty index is relatively cumbersome due to the large number of assessment items, some of which require patient cooperation. Subsequently, many researchers have continuously optimized and streamlined the items of the scale^9–11. However, not all of these scales have been validated, and the challenge of balancing clinical applicability with accuracy remains a point of contention.

According to a recent international consensus project concerning elderly patients with rectal cancer, frailty status has the potential to be utilized for risk stratification, targeted therapeutic strategies, and prognostic predictions of survival outcomes¹². The 5-item modified frailty index (mFI) is a concise and validated tool for assessing frailty status, with higher scores on the 5-item mFI being correlated with poorer clinical outcomes following trauma¹³, radical cystectomy¹⁴, and nephrectomy¹⁵. To date, there have been limited studies investigating the impact of preoperative frailty status on the clinical outcomes of elderly rectal cancer patients.

Thus, we aimed to evaluate the prevalence and adverse effects on outcomes of frailty in elderly patients undergoing resection for rectal cancer, as assessed by the 5-item mFI. We also sought to develop prediction models for risk stratification within this population.

Study Population

We conducted retrospective analyses of elderly patients undergoing surgical resection for rectal cancer at The First Affiliated Hospital of Xi’an Jiaotong University between January 2017 and December 2022. All patients provided informed consent for the use of their data. The main inclusion criteria were as follows: (1) patients aged 70 years or older; (2) patients diagnosed with rectal adenocarcinoma, confirmed by pathological examination post-surgery; (3) patients without other malignant diseases; and (4) patients who underwent curative resection. The exclusion criteria included the following: (1) patients who underwent endoscopic resection or surgical local resection; (2) patients who underwent secondary surgical resection due to local recurrence; (3) patients who underwent emergency surgery; and (4) patients with missing clinicopathological or follow-up data. The study protocol was approved by the Ethics Committee of The First Affiliated Hospital of Xi’an Jiaotong University (XJTU1AF2020LSL-004), and the study was conducted in accordance with the provisions of the Declaration of Helsinki.

Assessment of frailty status

we used 5-item mFI for the assessment of frailty status, which was developed and validated as previously prescribed^16–18. The 5-item mFI consists of 5 medical variables: (1) past severe chronic obstructive pulmonary disease (COPD), (2) congestive heart failure (CHF) in the month preceding surgery, (3) pre-surgery functional health status (independent vs. partially or totally dependent), (4) hypertension requiring medication, and (5) diabetes mellitus treated with oral agents or insulin. A score of 1 point was assigned for each item, with the total score varying from 0 to 5 points. The mFI was categorized as 0, 1, or ≥ 2, which was corresponded to favourable, moderate, and severe frailty status.

Data collection and patient followup

The following patient data were collected: preoperative assessment, operative characteristics, postoperative complications, postoperative recovery, and follow-up. Data were obtained from a prospective CRC medical record database. Postoperative complications were recorded according to the Clavien–Dindo (CD) classification system¹⁹. postoperative major complications were defined as CD grade ≥ II.

Follow-up of patients after discharge involved a clinical visit or telephone follow-up. Patients were reviewed in the outpatient clinic at 3 months interval during the first 2 years and every 6 months or annually for 5 years and then yearly. Telephone follow-up was conducted if an outpatient follow-up visit was not feasible. The last follow-up date in this study was 31 March 2024. The overall survival (OS) time was calculated from operation time to death or follow-up dead-line. Cancer-specific survival (CSS) was defined as the time from surgery to death due to rectal cancer.

Statistical analysis

Statistical analyses were conducted using IBM SPSS Statistics 26.0 software and R version 4.1. All statistical tests were two-sided at an α level of 0.05. Categorical variables were compared using the χ² test or Fisher’s exact test. Continuous variables were compared using the Mann–Whitney U test. Univariate and multivariate logistic analyses were used to evaluate the risk factors for postoperative major complications. The Kaplan–Meier method was used to analyze the survival data. Competing risk analysis was performed to find the associations between factors and rectal cancer-specific mortality. Rectal cancer mortality was regard as the interest event, while other causes-specific mortality was regard as the competing event. Based on the identified risk factors, predictive models for postoperative major complication and CSS were established. The receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA) were applied to evaluate the performance of predictive nomograms.

Patient demographics

From January 2017 to December 2022, a total of 1,614 elderly patients underwent radical resection for rectal cancer, of which 1,034 met the inclusion criteria. Among these, 318 patients (30.8%) were classified as having favorable frailty status (mFI = 0), 404 patients (39.1%) as moderate (mFI = 1), and 312 patients (30.1%) as severe (mFI ≥ 2). The mean age of the cohort was 76.1 years (± 4.1 SD), with 44.6% of the patients being female.

Table 1 summarizes the clinicopathological features according to mFI scores. Patients with an mFI ≥ 2 were found to be associated with older age (P < 0.001), higher American Society of Anesthesiology scores (P = 0.016), and a lower proportion of neoadjuvant chemoradiation history (P < 0.001). The most significant contributors to the mFI included hypertension (38.7%) and dependent functional status (23.9%), followed by diabetes (23.5%), COPD (14.9%), and CHF (2.9%).

Table 1

Clinical and pathological characteristics according to the 5-item modified frailty index in elderly rectal cancer patients
Characteristics		All patients (n = 1034)	Frailty Index = 0 (n = 318)	Frailty Index = 1 (n = 404)	Frailty Index ≥ 2 (n = 312)	p-value
Age, yr						<0.001
	70–74	480 (46.4)	194 (61.0)	197 (48.8)	89 (28.5)
	75–79	387 (37.4)	86 (27.0)	140 (34.7)	161 (51.6)
	80+	167 (16.2)	38 (12.0)	67 (16.5)	62 (19.9)
BMI, kg/m²						0.233
	<18.5	139 (13.4)	36 (11.3)	51 (12.6)	52 (16.7)
	18.5ཞ24.9	528 (51.1)	157 (49.4)	213 (52.7)	158 (50.6)
	25.0ཞ29.9	265 (25.6)	87 (27.4)	99 (24.5)	79 (25.3)
	≥ 30	102 (9.9)	38 (11.9)	41 (10.1)	23 (7.4)
Sex						0.142
	Male	573 (55.4)	173 (54.4)	213 (52.7)	187 (59.9)
	Female	461 (44.6)	145 (45.6)	191 (47.3)	125 (40.1)
ASA score						0.016
	I or II	521 (50.4)	178 (56.0)	204 (50.5)	139 (52.9)
	III or IV	513 (49.6)	140 (44.0)	200 (49.5)	173 (47.1)
History of COPD		232 (14.9)	0 (0.0)	70 (17.3)	162 (51.9)	<0.001
History of CHF		30 (2.9)	0 (0.0)	11 (2.7)	19 (6.1)	<0.001
Partially/totally dependent		247 (23.9)	0 (0.0)	102 (25.2)	141 (45.2)	<0.001
History of hypertension		400 (38.7)	0 (0.0)	132 (32.7)	268 (85.9)	<0.001
History of diabetes		243 (23.5)	0 (0.0)	89 (22.0)	154 (49.4)	<0.001
History of abdominal surgery		131 (12.7)	43 (13.5)	57 (14.1)	41 (13.1)	0.930
History of smoking		194 (18.8)	51 (16.0)	76 (18.8)	67 (21.5)	0.217
History of drinking		112 (10.8)	29 (9.1)	38 (8.7)	35 (11.2)	0.488
Neoadjuvant therapy						<0.001
	Yes	221 (21.4)	94 (29.6)	81 (20.0)	43 (13.8)
	No	813 (78.6)	224 (70.4)	323 (80.0)	173 (86.2)
Distance to the anal verge, cm		5.8 ± 2.2	5.5 ± 2.5	5.8 ± 2.3	6.0 ± 2.1	0.642
Tumor size, cm		3.7 ± 1.2	3.8 ± 1.3	3.7 ± 1.2	3.6 ± 1.1	0.749
Differentiation grade						0.133
	Well or moderate	905 (42.3)	279 (87.7)	362 (89.6)	187 (59.9)
	Poor or worse	129 (57.7)	39 (12.3)	42 (10.4)	48 (40.1)
Histology						0.349
	Adenocarcinoma	931 (90.0)	289 (90.9)	357 (88.4)	285 (91.3)
	Mucinous adenocarcinoma or signet-ring cell	103 (10.0)	29 (9.1)	47 (11.6)	27 (8.7)
Surgical approach						0.183
	Minimally invasive surgery	952 (92.1)	173 (93.4)	375 (92.8)	280 (89.7)
	Open surgery	82 (7.9)	21 (6.6)	29 (7.2)	32 (10.3)
p T stage						0.077
	T1-2	273 (26.4)	97 (30.5)	93 (23.0)	83 (26.6)
	T3-4	761 (73.6)	221 (69.5)	311 (77.0)	229 (73.4)
p N stage						0.453
	N0	607 (58.7)	194 (61.0)	238 (58.9)	175 (56.1)
	N1-2	427 (41.3)	124 (39.0)	166 (41.1)	137 (43.9)
p TNM stage						0.614
	I	197 (19.1)	61 (19.2)	83 (20.5)	53 (17.0)
	II	410 (39.6)	133 (41.8)	155 (38.4)	222 (39.1)
	III	427 (41.3)	124 (39.0)	166 (41.1)	137 (43.9)
Lymphovascular invasion						0.214
	Yes	258 (25.0)	84 (26.4)	89 (22.0)	85 (27.5)
	No	776 (75.0)	234 (73.6)	315 (78.0)	227 (72.8)
Perineural invasion						0.083
	Yes	303 (29.3)	90 (28.3)	107 (26.5)	106 (34.0)
	No	731 (70.7)	228 (71.7)	297 (73.5)	206 (66.0)
BMI, body mass index; ASA, American Society of Anesthesiology; COPD, chronic obstructive pulmonary disease; CHF, congestive heart failure.

Perioperative outcomes according to 5-item mFI scores

Table 2 presents the perioperative outcomes for patients categorized by their frailty status following surgery. Patients with higher mFI scores exhibited significantly elevated rates of postoperative overall complications (22.3% vs. 30.4% vs. 43.3%, P<0.001) and major complications (12.9% vs. 21.5% vs. 31.1%, P<0.001). When examining specific postoperative complications, the incidence of wound infection (6.6% vs. 13.1% vs. 18.9%, P<0.001), pulmonary infection (3.1% vs. 4.7% vs. 8.0%, P = 0.019), intestinal obstruction (3.8% vs. 3.7% vs. 7.7%, P = 0.027), and cardiovascular complications (1.3% vs. 1.5% vs. 4.5%, P = 0.010) was notably higher in patients with an mFI score of 2 or greater. Furthermore, in assessing postoperative recovery across three groups of patients, first postoperative exhaust time (2d vs. 3d vs. 4d, P<0.001), first eating liquid diet time (3d vs. 4d vs. 5d, P<0.001), and postoperative hospital time (9d vs. 10d vs. 12d, P<0.001) were significantly longer in in those classified as seriously debilitated. In addition, the total inpatient expenditure (74 thousand CNY vs. 7.8 thousand CNY vs. 8.1 thousand CNY, P = 0.042) was also significantly different among 3 groups.

Table 2

The perioperative outcomes in elderly rectal cancer patients stratified by the occurrence of 5-item modified frailty index.
Variables	Frailty Index = 0 (n = 318)	Frailty Index = 1 (n = 404)	Frailty Index ≥ 2 (n = 312)	p-value
Operation time (min), median (IQR)	205 (150–230)	200 (150–245)	215 (140–270)	0.376
Estimated blood loss (ml), median (IQR)	120 (50–180)	115 (60–170)	125 (55–185)	0.131
Postoperative complications (%)	71 (22.3)	123 (30.4)	135 (43.3)	<0.001
Clavien-Dindo grade ≥ II	41 (12.9)	87 (21.5)	97 (31.1)	<0.001
Surgical site infection	21 (6.6)	53 (13.1)	59 (18.9)	<0.001
Ileus	12 (3.8)	15 (3.7)	24 (7.7)	0.027
Anastomotic leakage	6 (1.9)	9 (2.2)	11 (3.5)	0.378
Anastomotic bleeding	7 (2.2)	5 (1.2)	8 (2.6)	0.405
Intra-abdominal abscess	5 (1.6)	11 (2.7)	9 (2.9)	0.494
Cardiovascular events	4 (1.3)	6 (1.5)	14 (4.5)	0.010
Pneumonia	10 (3.1)	19 (4.7)	25 (8.0)	0.019
Urinary infection	6 (1.9)	9 (2.2)	5 (1.6)	0.832
Urinary retention	11 (3.5)	6 (1.5)	10 (3.2)	0.188
Days to first flatus (days), median (IQR)	2 (1–4)	3 (1–4)	4 (1–5)	<0.001
Days to soft diet (days), median (IQR)	3 (2–5)	4 (2–5)	5 (2–6)	<0.001
Postoperative hospital stay (days), median (IQR)	9 (5–13)	10 (5–14)	12 (6–15)	<0.001
30-day readmission rate	11 (3.5)	16 (4.0)	14 (4.5)	0.804
Mortality	2 (0.6)	5 (1.2)	7 (2.2)	0.208
Total medical costs (thousand CNY), median (IQR)	74 (63, 85)	78 (66, 89)	81 (71, 90)	0.042
IQR, interquartile range; CNY, Chinese Yuan.

Unadjusted subgroup analyses by age group are shown in Supplementary Table 1. The mFI ≥ 2 was significantly associated with higher overall and major morbidity and total medical expenditure regardless of age group (all P<0.05).

Factors associated with postoperative major complications of elderly CRC patients

Table 3 displays the univariate and multivariable analyses of risk factors associated with postoperative major complications. In the univariate models, age ≥ 80 years, 5-item mFI ≥ 2, history of abdominal surgery, laparotomy and preoperative albumin count <35g/L were identified as risk factors for CD grade ≥ 2 postoperative complications (all P<0.05). Further multivariate logistic analysis results revealed that 5-item mFI ≥ 2 was associated with a significantly higher risk of postoperative major complications compared to a score of 1 or 0 [odds ratio (OR) = 3.145, 95% confidence interval (CI): 2.047–4.833, P<0.001]. In addition, history of abdominal surgery (OR = 1.719, 95% CI: 1.114–2.652, P = 0.014), laparotomy (OR = 4.967, 95% CI: 3.059–8.065, P<0.001), and preoperative albumin count <35g/L (OR = 2.258, 95% CI: 1.646–3.099, P<0.001) were also identified as independent risk factors for CD grade ≥ 2 postoperative complications.

Table 3

Multivariate analyses of risk factors for postoperative major complications in elderly rectal cancer patients after curative surgery.
Variables			Univariate analysis		Multivariate analysis
Variables			Odds ratio (95 CI)	p-value	Odds ratio (95 CI)	p-value
Age				0.042		0.430
	70–74		Reference		Reference
	75–79		1.164 (0.837–1.620)		1.075 (0.758–1.526)
	80+		1.626 (1.083–2.440)		1.337 (0.862–2.075)
Sex				0.615
	male		Reference
	female		0.926 (0.688–1.248)
BMI				0.460
	<18.5		Reference
	18.5ཞ24.9		0.953 (0.601–1.511)
	25.0ཞ29.9		1.233 (0.751–2.014)
	≥ 30		1.232 (0.670–2.264)
ASA score				0.063
	I or II		Reference
	III or IV		1.326 (0.985–1.783)
5-item modified frailty index				<0.001		<0.001
	0		Reference		Reference
	1		1.854 (1.237–2.779)		1.873 (1.226–2.860)
	≥ 2		3.048 (2.030–4.576)		3.145 (2.047–4.833)
Tumor size, mm				0.531
	<40		Reference
	>40		0.910 (0.677–1.223)
History of abdominal surgery				0.010		0.014
	No		Reference		Reference
	Yes		1.706 (1.137–2.559)		1.719 (1.114–2.652)
History of smoking				0.296
	No		Reference
	Yes		1.213 (0.844–1.743)
History of drinking				0.608
	No		Reference
	Yes		1.135 (0.700-1.842)
Distance to the anal verge, cm				0.681
	11–15		Reference
	6–10		1.167 (0.810–1.681)
	0–5		1.034 (0.719–1.489)
Differentiation grade				0.812
	Well or moderate		Reference
	Poor or worse		1.055 (0.682–1.632)
Histology				0.259
	Adenocarcinoma		Reference
	Mucinous adenocarcinoma or signet-ring cell		0.702 (0.380–1.298)
Surgical approach				<0.001		<0.001
	Minimally invasive surgery*		Reference		Reference
	Open surgery		5.216 (3.279–8.298)		4.967 (3.059–8.065)
Clinical T stage				0.283
	T1-2		Reference
	T3-4		1.219 (0.849–1.751)
Clinical N stage				0.492
	N0		Reference
	N1-2		1.110 (0.824–1.494)
Neoadjuvant therapy				0.185
	No		Reference
	Yes		1.341 (0.869–2.070)
Preoperative WBC count, 10⁹/L				0.730
		<10	Reference
		≥ 10	1.053 (0.784–1.416)
Preoperative albumin count, g/L				<0.001		<0.001
		≥ 35	Reference		Reference
		<35	2.197 (1.628–2.996)		2.258 (1.646–3.099)
Preoperative CEA level, µg/L				0.230
		<5	Reference
		≥ 5	1.236 (0.875–1.748)
Preoperative CA19-9 level, µg/L				0.491
		<37	Reference
		≥ 37	1.162 (0.757–1.784)
ASA: American society of Aneshesiologists; *: Laparoscopic surgery or robotic surgery; WBC: white blood cell.

To further improve the clinical practicability, the results of multivariate analysis were used to build the postoperative major complication nomogram (Fig. 1A). The area under the curve (AUC) of this nomogram was 0.730 (95% CI, 0.691–0.769), and calibration curve were consistent with the observation results (Figs. 1B, 1C). Additionally, the DCA curves illustrated that the nomogram had good clinical utility (Fig. 1D).

Impact of 5-item mFI on survival of elderly rectal cancer patients

A significant association was found between different frailty states and prognosis of elderly rectal patients (Fig. 2). Compared to the 5-year OS of patients with an mFI of 1 (53.5%) and 0 (67.6%), the 5-year OS of patients with an mFI of ≥ 2 exhibited a markedly lower 5-year OS of only 44.4% (P<0.001). Similar results were obtained for 5-year CSS. The cumulative 5-year incidence of CSS in patients with severe frailty was 57.1%, demonstrating significantly improved outcomes compared to those with moderate (71.4%) and favorable frailty (76.8%). Furthermore, we conducted a subgroup survival analysis based on 5-item mFI scores across different tumor stages. The prognosis of various frailty states in elderly patients with rectal cancer varies significantly across different tumor stages, except for the CSS of patients at stage III (Supplementary Figs. 1–3).

Factors associated with CSS of elderly rectal cancer patients

The competing risk model results for CSS are shown in Table 4. Results from the univariate analyses demonstrated that age, 5-item mFI, differentiation grade, T stage, N stage, TNM stage, lymphovascular invasion, perineural invasion, and adjuvant therapy were identified as risk factors for CSS of elderly rectal cancer patients (all P<0.05). TNM stage was not included in the multiple regression models due to strong collinearity with three categories of N stage. Multivariate competing risk analyses indicated that the factors of age ≥ 80 years [subdistribution hazard ratio (SHR) = 1.56, 95% CI:1.12–2.17, P = 0.008], 5-item mFI ≥ 2 (SHR = 2.00, 95% CI: 1.47–2.72, P<0.001), poorly differentiated or undifferentiated (SHR = 1.52, 95% CI: 1.16–2.99, P<0.001), T4 stage (SHR = 3.51, 95% CI: 1.94–6.35, P<0.001), N2 stage (SHR = 2.49, 95% CI: 1.69–3.68, P<0.001), lymphovascular invasion (SHR = 1.91, 95% CI: 1.27–2.88, P = 0.002), and adjuvant therapy (SHR = 0.66, 95% CI: 0.46–0.97, P = 0.032) were independent predictors of CSS in elderly rectal cancer patients.

Table 4

Multivariate competing risk analyses of risk factors for cancer specific survival in elderly rectal cancer patients after curative surgery.
Variables			Univariate analysis		Multivariate analysis
Variables			SHR (95% CI)	p-value	SHR (95% CI)	p-value
Age
	70–74		Reference		Reference
	75–79		1.07 (0.78–1.45)	0.690	1.01 (0.74–1.38)	0.940
	80+		1.64 (1.21–2.24)	0.001	1.56 (1.12–2.17)	0.008
Sex				0.623
	male		Reference
	female		0.94 (0.75–1.19)
BMI
	<18.5		Reference
	18.5ཞ24.9		0.91 (0.64–1.28)	0.591
	25.0ཞ29.9		1.36 (0.99–1.87)	0.064
	≥ 30		1.02 (0.72–1.43)	0.933
ASA score				0.514
	I or II		Reference
	III or IV		0.918 (0.71–1.19)
5-item modified frailty index
	0		Reference		Reference
	1		1.62 (1.19–2.20)	0.002	1.76 (1.28–2.42)	<0.001
	≥ 2		2.06 (1.52–2.80)	<0.001	2.00 (1.47–2.72)	<0.001
Tumor size, mm				0.192
	<40		Reference
	>40		1.19 (0.92–1.53)
History of abdominal surgery				0.753
	No		Reference
	Yes		1.04 (0.81–1.34)
History of smoking				0.156
	No		Reference
	Yes		1.03 (0.91–1.86)
History of drinking				0.953
	No		Reference
	Yes		0.98 (0.57–1.70)
Distance to the anal verge, cm
	11–15		Reference
	6–10		1.03 (0.76–1.40)	0.842
	0–5		1.14 (0.84–1.55)	0.393
Differentiation grade				<0.001		<0.001
	Well or moderate		Reference		Reference
	Poor or worse		1.73 (1.33–2.24)		1.52 (1.16–1.99)
Histology				0.854
	Adenocarcinoma		Reference
	Mucinous adenocarcinoma or signet-ring cell		0.97 (0.72–1.31)
Surgical approach				0.174
	Minimally invasive surgery*		Reference
	Open surgery		0.83 (0.65–1.08)
p T stage
	T1		Reference		Reference
	T2		1.45 (0.79–2.66)	0.240	1.25 (0.68–2.29)	0.470
	T3		2.43 (1.42–4.17)	0.001	1.54 (0.89–2.68)	0.130
	T4		5.99 (3.38–10.60)	<0.001	3.51 (1.94–6.35)	<0.001
p N stage
	N0		Reference		Reference
	N1		2.23 (1.66–2.99)	<0.001	1.67 (1.19–2.32)	0.003
	N2		3.35 (2.34–4.80)	<0.001	2.49 (1.69–3.68)	<0.001
p TNM stage
	I		Reference
	II		2.11 (1.45–3.07)	<0.001
	III		4.09 (2.02–5.92)	<0.001
Neoadjuvant therapy				0.147
	No		Reference
	Yes		0.83 (0.64–1.06)
Lymphovascular invasion				<0.001		0.002
		Yes	Reference		Reference
		No	1.90 (1.43–2.54)		1.91 (1.27–2.88)
Perineural invasion				0.002		0.590
		Yes	Reference		Reference
		No	1.89 (1.26–2.84)		1.11 (0.76–1.61)
Adjuvant therapy				0.045		0.032
	No		Reference		Reference
	Yes		0.69 (0.47–0.99)		0.66 (0.46–0.97)
Preoperative WBC count, 10⁹/L				0.857
	<10		Reference
	≥ 10		0.97 (0.72–1.31)
Preoperative albumin count, g/L				0.584
	<35		Reference
	≥ 35		0.89 (0.60–1.32)
Preoperative CEA level, µg/L				0.372
	<5		Reference
	≥ 5		1.12 (0.87–1.45)
Preoperative CA19-9 level, µg/L				0.127
	<37		Reference
	≥ 37		1.32 (0.95–1.84)
SHR: subdistribution hazard ration; ASA: American society of Aneshesiologists; *: Laparoscopic surgery or robotic surgery; WBC: white blood cell.

The CSS prediction nomogram of the elderly rectal cancer patients was further constructed according to the multivariate competing risk analysis results (Fig. 3). The AUCs of 1-, 3-, and 5-year CSS of nomogram were 0.705 (95% CI: 0.639–0.771), 0.791 (95% CI: 0.751–0.832), and 0.844 (95% CI: 0.806–0.898), which was statistically superior to that of TNM staging systems. Both the calibration curves and DCA curves demonstrated that incorporating frailty status and clinicopathological characteristics into predictive model significantly enhanced the accuracy of predicting CSS (Supplementary Figs. 4–6).

As the global population ages, there has been an increasing emphasis on developing management strategies for older patients with CRC who are facing debilitating conditions. This study is the first to explore the impact of the 5-item mFI on clinical outcomes for elderly patients with rectal cancer. Our findings demonstrated that preoperative frailty status is associated with a higher incidence of postoperative complications, prolonged recovery times, increased medical costs, and worse cancer-specific mortality among elderly patients with rectal cancer. Furthermore, we constructed a nomogram for predicting major postoperative complications and a CSS prediction nomogram, both of which can be utilized to stratify risk and guide clinical decision-making.

With the increasing attention paid to frailty, numerous frailty assessment tools have been developed and preliminarily validated. These tools can be categorized into two main types: frailty phenotype models and frailty accumulation models²⁰. The frailty phenotype model emphasizes overall physical decline, where grip strength, walking speed, physical activity, body weight, or one-dimensional deficiencies are combined to define the individual's frailty status. Typical representatives of this model include the Aging and Retirement in Europe scale²¹ and Frailty Phenotype scale²². Conversely, the frailty accumulation model defines frailty as the ratio of the number of deficits present to the total number of items considered, focusing more on the accumulation of deficits. A key representative of this model is the frailty index, which encompasses 70 factors, including cognitive function, nutritional status, depression, social support, and comorbidities; a score exceeding 0.25 is indicative of the debilitating condition⁸. Although the frailty index has proven effective in predicting morbidity and mortality among patients, many items within the frailty index are challenging to assess and implement in real clinical environments. An effective frailty assessment tool should not only objectively differentiate between frailty categories but also possess high clinical sensitivity and easy to operate. Consequently, the frailty index has been redefined and abbreviated across various databases and studies. Searle et al. optimized the frailty index to 11 items using data from the National Surgical Quality Improvement Program study, maintaining close predictive efficiency²³. As a continuous scale measure, the 11-item mFI offers good sensitivity even at the lower end of the frailty continuum, thereby facilitating research involving elderly individuals. Later, the study by Subramaniam et al. demonstrated that both the 5-item mFI and the 11-item mFI are equally effective predictors in surgical patients and the 5-point mFI is a strong predictor of mortality and postoperative complications²⁴. Therefore, in this study 5-item mFI was chose as a preoperative frailty assessment tool for elderly patients with rectal cancer.

The results of our study showed that increased odds of postoperative overall and major morbidity for patients with higher 5-item mFI scores in both overall and stratification analyses. Okabe et al. used the clinical frailty scale (CFS) to evaluate the frailty of CRC patients older than 65 years and found that CFS scores ≥ 4 is a reliable predictor of postoperative severe complications²⁵. In the meta-analysis by Moreno-Carmona et al. shown that preoperative frailty increased the risk of overall (OR = 2.34, 95% CI: 1.75–3.15, P<0.001) and severe (OR = 2.443, 95% CI: 1.72–3.43, P<0.001) complications in patients with colon cancer²⁶. The results of our study were consistent with previous research. The 5-item mFI score, which encompasses multiple factors such as COPD, CHF, hypertension, and diabetes, effectively reflects patients' functional reserve and physiological response to surgical stress. Additionally, the 5-point mFI includes an independent status assessment that may influence early postoperative mobilization and ambulation of patients. Early ambulation following surgery can enhance the recovery of gastrointestinal function, prevent postoperative pulmonary complications, and reduce the length of hospital stay²⁷. In our study, we further developed a nomogram to quantify the risk of postoperative morbidity, which can be utilized for risk stratification and developed individualized rehabilitation strategies to reduce the risk of postoperative complications.

In our study, preoperative frailty status was confirmed as an independent risk factor for cancer-specific mortality (SHR = 2.00, 95% CI: 1.47–2.72, P<0.001) in elderly patients with rectal cancer, as demonstrated by competing risk models. The research conducted by Ugolini et al. utilized the Groningen Frailty Index as the frailty assessment tool, revealing a significant association between frailty status and OS in CRC patients aged over 70 years²⁸. Additionally, a meta-analysis published by Chen et al.²⁹, identified a link between frailty and CSS (OR = 4.60, 95% CI: 2.75–7.67, P < 0.001) of CRC patients. The results of our study were consistent with those of the aforementioned studies. The mechanisms underlying the relationship between frailty status and prognosis are likely multifactorial. Firstly, previous studies have indicated that cancer patients exhibiting frailty are thought to have altered responses to chemotherapy, along with increased toxicity^{30, 31}. Moreover, the notably lower proportion of debilitated patients with rectal cancer receiving standard postoperative adjuvant therapy may also contribute to the poor prognosis observed in this population. Secondly, investigations into frailty and immune status have demonstrated that the prevalence of frailty in older cancer patients is significantly associated with inflammatory cytokines, including C-reactive protein, IL-6, and TNF-α^{32, 33}. Thus, altered immune status may play a crucial role in the pathogenesis of frailty, either directly or through its detrimental effects on other physiological systems, leading to poorer prognoses. Lastly, our study found that preoperative frailty status was significantly associated with severe postoperative complications, which have been confirmed to impact the long-term prognosis of patients³⁴.

Preoperative frailty status assessment not only contributes to patient risk stratification, but more importantly, helps in the design and implementation of treatment strategies. According to the expert consensus, preoperative debilitating condition can be targeted intervention to alleviating or even reversing the symptoms. Recently, many exploratory studies have demonstrated the value of preoperative prehabilitation schedules for improving frailty status of patients with surgery^{35, 36}. Ommundsen et al. compared the perioperative clinical outcome between CRC patients receiving tailored care following a geriatric assessment and patients receiving standard care, through a randomized controlled study³⁷. Secondary analyses showed that the overall incidence of postoperative complications (CD I–V) was significantly lower in the intervention group (OR = 0.33, 95% CI: 0.11–0.95, P = 0.05). In the PREHAB randomized clinical trial, patients in the experimental group received a 4-week multimodal rehabilitation program, including a high-intensity exercise program, nutritional intervention, psychological support, and a smoking cessation program³⁸. According to intention-to-treat analysis, patients who received rehabilitation program had significantly lower severe (OR = 0.47, 95% CI: 0.26–0.87, P = 0.02) and medical (OR = 0.448, 95% CI: 0.26–0.89, P = 0.02) complications. However, in the study by Li et al. ³⁹ and Burden et al. ⁴⁰, found that rehabilitation programs had no significant effect on improving the incidence of postoperative complications. Thus, Pre-rehabilitation programs for CRC patients are still controversial. Much of the controversy surrounding the period of pre-rehabilitation program, the dimension of pre-rehabilitation schedule, how to intervene, and so on. Therefore, when formulating a patient's early rehabilitation plan, it is necessary to fully consider the patient's clinical characteristics, select appropriate frailty assessment tools, and the implementation of interventions.

This study has several limitations. Firstly, it is a retrospective, single-center design; therefore, a multicenter prospective study with a larger sample size is necessary to validate the conclusions drawn from this research. Secondly, this study focuses solely on postoperative major complications and cancer-specific mortality for further risk assessment, excluding other specific clinical outcomes such as anastomotic leakage, anastomotic bleeding, and disease-free survival from the analysis. Thirdly, although we controlled for confounders in all analyses, some bias may still be present due to unobserved variables. Lastly, a patient's frailty status is subject to constant change over time, and this study did not further assess the impact of these dynamic changes in frailty status on prognosis following surgery.

In conclusion, 5-item mFI is a simple and effective tool for assessing frailty in elderly patients with rectal cancer. Higher 5-item mFI scores were associated with worse short-term and long-term clinical outcomes of patients after surgical resection. Furthermore, integrating preoperative frailty assessment with clinicopathological characteristics may improve risk stratification in older patients with rectal cancer, further facilitating clinical decision-making and bringing prognostic benefits to patients.

Disclosures

The authors declare no conflicts of interest.

Funding information:

This project was supported by National Natural Science Foundation of China (No. 81870380) and Shaanxi Province Science Foundation (S2024-YF-YBSF-1247).

Author Contribution

Conception/design: H.W., J.S. Provision of study material or patients: H.W., F.S., C.H., L.Z. Collection and/or assembly of data: H.W., C.H. Data analysis and interpretation: P.Q., C.H., L.Z. Final approval of manuscript: H.W., F.S., C.H., L.Z., P.Q., J.S.

Data Availability

Requests for raw data and supporting materials can be made by contacting the corresponding author. We can willingly provide anonymized patient data via the editorial office. The corresponding author's correspondence address and email address are: No. Address: No. 277, Yanta West Road, Xi'an City, Shaanxi Province, China, Postal Code: 710061, Email address: [email protected].

Sung H. et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 71(3):209–249 (2021).
Cohen C.I. et al. Frailty: A Multidimensional Biopsychosocial Syndrome. Med Clin North Am. 107(1):183–197 (2023).
Dent E. et al. Management of frailty: opportunities, challenges, and future directions. Lance. 394(10206):1376–1386 (2019).
Hoogendijk E.O. et al. Frailty: implications for clinical practice and public health. Lancet. 394(10206):1365–1375 (2019).
A C P. et al. Impact of frailty on outcomes in surgical patients: A systematic review and meta-analysis. Am J Surg, 218(2):393–400 (2018).
Alberto P. et al. A multidimensional approach to frailty in older people. Ageing Res Rev, 60(0) (2020).
Komici K. et al. Frailty in Patients With Lung Cancer: A Systematic Review and Meta-Analysis. Chest, 162(2) (2022).
Rockwood K. et al. Prevalence, attributes, and outcomes of fitness and frailty in community-dwelling older adults: report from the Canadian study of health and aging. J Gerontol A Biol Sci Med Sci. 59(12):1310–1317 (2004).
Palliyaguru D.L. et al. Frailty index as a biomarker of lifespan and healthspan: Focus on pharmacological interventions. Mech Ageing Dev. 180:42–48 (2019).
Tanaka T. et al. Oral Frailty Index-8 in the risk assessment of new-onset oral frailty and functional disability among community-dwelling older adults. Arch Gerontol Geriatr. 94:104340 (2021).
Khanna A.K. et al. Automated Electronic Frailty Index-Identified Frailty Status and Associated Postsurgical Adverse Events. JAMA network open. 6(11):e2341915 (2023).
Podda M. et al. Multidisciplinary management of elderly patients with rectal cancer: recommendations from the SICG (Italian Society of Geriatric Surgery), SIFIPAC (Italian Society of Surgical Pathophysiology), SICE (Italian Society of Endoscopic Surgery and new technologies), and the WSES (World Society of Emergency Surgery) International Consensus Project. World J Emerg Surg. 16(1):35 (2021).
Guiab K. et al. Can the 5-item Modified Frailty Index Predict Outcomes in Geriatric Trauma? A National Database Study. World J Surg. 46(10):2328–2334 (2022).
Yamashita S. et al. Modified 5-Item Frailty Index Score as Prognostic Marker After Radical Cystectomy in Bladder Cancer. Clin Genitourin Cancer. 20(3):e210-e216 (2022).
Goldwag J. et al. 5-Item Modified Frailty Index as a Preoperative Predictor of Morbidity Following Minimally Invasive Partial Nephrectomy. Urology. 157:138–142 (2021).
Chen S.Y. et al. The Effect of Frailty Index on Early Outcomes after Combined Colorectal and Liver Resections. J Gastrointest Surg. 22(4):640–649 (2018).
Farhat J.S. et al. Are the frail destined to fail? Frailty index as predictor of surgical morbidity and mortality in the elderly. J Trauma Acute Care Surg. 72(6):1526–1530; discussion 1530 – 1521 (2012).
Uppal S. et al. Frailty index predicts severe complications in gynecologic oncology patients. Gynecol Oncol. 137(1):98–101 (2015).
Clavien P.A. et al. The Clavien-Dindo classification of surgical complications: five-year experience. Ann Surg. 250(2):187–196 (2009).
Howlett S.E. et al. The degree of frailty as a translational measure of health in aging. Nat Aging. 1(8):651–665 (2021).
Wagner M. et al. [Distance caregiving: Empirical findings from the Survey of Health, Ageing and Retirement in Europe]. Z Gerontol Geriatr. 52(6):529–536 (2019).
Musso C.G. et al. Frailty phenotype and chronic kidney disease: a review of the literature. Int Urol Nephrol. 47(11):1801–1807 (2015).
Searle S.D. et al. A standard procedure for creating a frailty index. BMC geriatrics. 8:24 (2008).
Subramaniam S. et al. New 5-Factor Modified Frailty Index Using American College of Surgeons NSQIP Data. J Am Coll Surg. 226(2):173–181.e178 (2018).
Okabe H. et al. Frailty predicts severe postoperative complications after elective colorectal surgery. Am J Surg. 217(4):677–681 (2019).
Moreno-Carmona M.R. et al. Effect of frailty on postoperative complications, mortality, and survival in older patients with non-metastatic colon cancer: A systematic review and meta-analysis. J Geriatr Oncol. 15(2):101639 (2024).
Forsmo H.M. et al. Enhanced Recovery After Colorectal Surgery (ERAS) in Elderly Patients Is Feasible and Achieves Similar Results as in Younger Patients. Gerontol Geriatr Med. 3:2333721417706299 (2017).
Ugolini G. et al. How to select elderly colorectal cancer patients for surgery: a pilot study in an Italian academic medical center. Cancer Biol Med. 12(4):302–307 (2015).
Chen S. et al. Frailty and long-term survival of patients with colorectal cancer: a meta-analysis. Aging Clin Exp Res. 34(7):1485–1494 (2022).
Hay C.M. et al. Chemotherapy in older adult gynecologic oncology patients: Can a phenotypic frailty score predict tolerance? Gynecol Oncol. 152(2):304–309 (2019).
Akbıyık I. et al. Association of frailty with chemotherapy toxicity in chemotherapy-naive older patients with cancer. Support Care Cancer. 30(12):9975–9981 (2022).
Yao X. et al. Inflammation and immune system alterations in frailty. Clin Geriatr Med. 27(1):79–87 (2011).
Harneshaug M. et al. Screening for frailty among older patients with cancer using blood biomarkers of inflammation. J Geriatr Oncol. 10(2):272–278 (2019).
Fransgaard T. et al. The impact of postoperative complications and delay of adjuvant chemotherapy on oncological outcomes in patients with colorectal cancer. Colorectal Dis. 23(5):1132–1140 (2021).
Ven Fong Z. et al. Contemporary Opportunity for Prehabilitation as Part of an Enhanced Recovery after Surgery Pathway in Colorectal Surgery. Clin Colon Rectal Surg. 32(2):95–101 (2019).
Cuijpers A.C.M. et al. Quality and clinical generalizability of feasibility outcomes in exercise prehabilitation before colorectal cancer surgery - A systematic review. Eur J Surg Oncol. 48(7):1483–1497 (2022).
Ommundsen N. et al. Preoperative geriatric assessment and tailored interventions in frail older patients with colorectal cancer: a randomized controlled trial. Colorectal Dis. 20(1):16–25 (2018).
Molenaar C.J.L. et al. Effect of Multimodal Prehabilitation on Reducing Postoperative Complications and Enhancing Functional Capacity Following Colorectal Cancer Surgery: The PREHAB Randomized Clinical Trial. JAMA Surg. 158(6):572–581 (2023).
Li C. et al. Impact of a trimodal prehabilitation program on functional recovery after colorectal cancer surgery: a pilot study. Surg Endosc. 27(4):1072–1082 (2013).
Burden S.T. et al. An unblinded randomised controlled trial of preoperative oral supplements in colorectal cancer patients. J Hum Nutr Diet. 24(5):441–448 (2011).

No competing interests reported.

Supplementarymaterials.docx

The effect of 5-item modified frailty index on clinical outcomes in elderly rectal cancer patients after curative surgery

Status:

Version 1

Abstract

Figures

Introduction

Methods