A nomogram for the prediction of overall survival and Prognostic significance of different treatment modalities for elderly Hepatocellular carcinoma

doi:10.21203/rs.3.rs-3395904/v1

Download PDF

Research Article

A nomogram for the prediction of overall survival and Prognostic significance of different treatment modalities for elderly Hepatocellular carcinoma

https://doi.org/10.21203/rs.3.rs-3395904/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background：Hepatocellular carcinoma (HCC) has a poor prognosis, especially in elderly HCC patients. A growing number of studies have confirmed that nomogram based on demographic information, pathologic information, and treatment modalities are effective in predicting the prognosis of cancer patients. We aimed to develop an individualized and clinically useful nomogram for elderly patients.We also aimed to investigate the prognostic significance of different treatment modalities.
Method：A total of 7694 eligible elderly patients with HCC were identified from the Surveillance, Epidemiology, and End Results (SEER) database between 2000 and 2020. The patients were randomized into the training cohort (n = 5385) and the validation cohort (n = 2309) at a 7:3 ratio. Univariate and multivariate COX regression analyses were performed in the training cohort. Based on the results of the regression analysis, a nomogram model for predicting the prognosis of elderly HCC patients was established and verified in the validation cohort. The performance of the nomogram was evaluated by using receiver operating characteristic curve (ROC) and calibration curve.Finally, we grouped patients by surgery, chemotherapy, radiotherapy separately, and then used Propensity Score Matching (PSM) to balance other variables to explore prognostic significance of different treatment modalities for elderly Hepatocellular carcinoma.

Results: Regression analysis showed that Age, Grade, T.stage, N.stage, M.stage, Chemotherapy, Radiotherapy, and Surgery had significant prognosis and developed nomogram based on the above 8 variables.This nomogram showed good accuracy in predicting one-year OS of the training cohort (AUC=0.745) and was confirmed in the validation cohort (AUC=0.786). The calibration curve shows that the actual survival probability and the predicted survival probability are in good agreement. A risk classification system was established to risk-stratify elderly patients with HCC,ROC was used to validate the efficacy of nomogram model in differentiating between high and low risk groups(p<0.01) ,and verify in the validation cohort(p<0.01).Finally, after grouping elderly HCC patients according to different treatment modalities, subgroup analysis showed that:The survival time of the surgery group was longer than that of the no-surgery group. Median overall survival was longer in the chemotherapy group than in the no-chemotherapy group, but there was no advantage in long-term prognosis. Median overall survival was also significantly longer in the radiotherapy group than in the no-radiotherapy group, but when analyzed in subgroups, it was found that in the T4 stage and GradeIV subgroups, radiotherapy patients had a worse prognosis than those who did not receive radiotherapy.

Conclusion: The nomogram established based on demographic characteristics and other clinical parameters can be used as an effective tool to predict the prognosis of elderly HCC patients, and its good prediction efficiency can realize the individualized treatment plan formulation and follow-up management of elderly HCC patients.

Hepatocellular carcinoma

nomogram

overall survival

Prognostic significance

treatment modalities

Primary liver cancer (PLC) is the major cause of cancer mortality globally and ranks 3rd among cancer-related mortality causes(Weir, Thompson et al. 2015, Sung, Ferlay et al. 2021).Hepatocellular carcinoma (HCC) is the most common pathologic type of PLC, accounting for 85–90%(Bruix, Gores et al. 2014).In light of the current global aging trend, we focused on the elderly HCC population, which, like many cancers, has a worse prognosis.Currently, the TNM staging system released by the American Joint Committee on Cancer (AJCC) remains the primary method for assessing disease progression, treatment options, and prognosis in HCC, but this staging system contains fewer clinical parameters to measure the prognosis of HCC patients and is applicable to any cancer population, which is not sufficiently specific to predict the prognosis of HCC, especially in elderly HCC patients.Recent years, many studies have demonstrated that marital status, family income, etc. have a vital impact on the prognosis of patients with cancer(Xu, Liu et al. 2017, Cai, Fan et al. 2020, Chen, Yin et al. 2020).Some studies have also proved portal vein tumor thrombus(PVTT), an early phenomenon of vascular invasion in hepatocellular carcinoma, seriously affects the prognosis of hepatocellular carcinoma(Fan, Wu et al. 2001, Wei, Jiang et al. 2019, Su, Gu et al. 2022).Some studies have also proved portal vein tumor thrombus, an early phenomenon of vascular invasion in hepatocellular carcinoma, seriously affects the prognosis of hepatocellular carcinoma(Li, Feng et al. 2016).Therefore, these factors are included in this article in predicting the prognosis of elderly patients with hepatocellular carcinoma.This article is divided into two parts.The first part is to develop a prognostic nomogram model more applicable for elderly HCC patients and to validate the predictive efficacy of the model.In the second part, we consider that the treatment of HCC patients is determined by the physician's choice based on the patient's condition,that is, the decision of a patient to undergo surgery/radiotherapy/chemotherapy is the result of the physician's careful selection of the patient.This non-randomized bias may affect the comparison of survival among patients, so we utilizes Propensity Score Matching to control for potential confounders and further more accurately compare the survival differences of patients under different treatment modalities,which can help in the selection of optimal clinical treatment decisions for patients.Therefore, this two-part study aims to develop an individualized and more clinically practical nomogram model for elderly patients, which will help in prognostic prediction, therapeutic decision-making and follow-up management of these patients.

2.1 Patients cohort

The research analysis in this article was conducted based on the SEER database using SEER.Stat to extract basic information about HCC patients from 17 registries in the United States from 1 January 2000- 31 December 2020, including patient demographics, tumor characteristics, diagnosis, treatment modalities, survival status, and income level.All patients were classified according to the American Joint Committee on Cancer/Union for International Cancer Control (AJCC/UICC) 6th edition tumor-lymph node-metastasis (TNM) staging system.Information of the Portal vein tumor thrombus (PVTT) is available according to CS.extension and N.stage within the SEER database(Tortora, Farella et al. 2023).Overall survival (OS) expressed in months (time from diagnosis to death from any cause.Since the patient data extracted from SEER database were de-identified and available to the public for research purposes, informed patient consent was not required for this study and ethical approval was waived by the local ethics committee.The inclusion criteria were as follows:(1)Hepatocellular carcinoma is the primary tumor,International Classification of Diseases codes 8170/3, 8171/3, 8172/3, 8173/3, 8174/3, 8175/3 are used to identify patients with HCC.The exclusion criteria are as follows:(1) Age < 65 years;(2)Diagnosis not confirmed microscopically.(3)Unknown race, unknown degree of tumor differentiation, unknown TNM stage of the tumor, unknown whether surgery was performed, unknown size of the tumor, unknown degree of tumor invasion, and Tx stage,Nx stage,Mx stage and unknown survival time were not included in the nomogram.Finally,7694 eligible patients were gathered for analysis.Then the training cohort (n = 5385) and validation cohort (n = 2309) were generated using R software in randomized groups in a 7:3 ratio.The complete data screening process was shown in Fig. 1.

2.2 Statistical analysis

SPSS 26 and R software 4.3.1 were used for statistical analysis.All variables are further transformed into categorical variables and expressed in frequency and proportion. Chi-square tests were used to compare differences in baseline data between groups.Considering that the database collects information on cancer patients in the United States, there exists some bias in the variable of race, so race information was not used to develop the regression formula.Variables with statistically significant results from COX regression analyses were included in the establishment of prognostic nomogram predicting the probability of OS (p < 0.01 considered statistically significant).The 'rms' R package was used to construct the nomogram, Kaplan-Meier were used to compare survival probilitity between different subgroups, and log rank tests were used to compare differences between groups.The performance of the nomogram was evaluated by calculating area under the receiver operating characteristic curve(AUC).The greater the value of AUC, the greater the validity of the nomogram. AUC > 0.7 is considered acceptable. Calibration curve with 1,000 resamplings were applied to compare predicted and observed survival rates.Each patient in the validation cohort was scored according to the nomogram and the patients were categorized into low-risk group and high-risk group according to their scores using the dichotomy,then using ROC to validate the efficacy of nomogram in differentiating between high and low risk groups(p < 0.01),and verify in the validation cohort(p < 0.01).

2.3 Propensity Score-Matching

Propensity Score Matching (PSM) was utilized to minimize the effects of selection bias and confounding variables.PSM was implemented using MatchIt, tableone packages in R4.3.1 software.The scores were calculated according to the Logistic regression model, and matched in equal proportions by using the method of near neighbor matching. The difference between the baseline characteristics of the two groups was set as a parameter of 1:1 matching, and the correction threshold was set at 0.02, and the validation of PSM was realized by comparing the differences of each variable before and after PSM. Categorical variables were compared using the X²test.

3.1 Demographic and clinical characteristics

A total of 7694 patients were enrolled in the study and randomly divided into a training cohort (n = 5385) and a validation cohort (n = 2309) at a ratio of 7:3.The demographic and clinicopathological data of the training cohort and validation cohort are shown in Table 1, Of all patients included in this study, 5433 (70.6%) were male and 3265 (42.4%) were female; 4429 (57.6%) were 65–75 years old and 3265 (42.4%) were ≥ 75 years old; 611 (7.9%) were blacks, 5381 (69.9%) were whites, and 1702 (22.1%) were of other races ; 4838 (62.9%) had tumors < 5 cm in diameter, and a total of 2856 (37.1%) had tumors ≥ 5 cm in diameter; in terms of the degree of tumor differentiation, there were a total of 6029 (79.2%) of moderately and highly differentiated, and a total of 1602 (20.8%) of poorly differentiated and undifferentiated; according to the ATCC staging, there were 3915 (50.9%) in T1 stage, 1651 (21.5%) in T2 stage (21.5%), 1857 (24.1%) in T3 stage, 271 (3.5%) in T4 stage; 7309 (95.0%) in N0 stage, 7000 (91.0)% in M0 stage, and 537 (7.0%) patients with PVTT; in terms of treatments: a total of 558 (7.3%) were treated with radiotherapy, 2577 (33.5%) with chemotherapy, and 3441 (44.7%) received surgery, and in terms of family income, 3896 (50.6%) from the middle class accounted for the majority of cases.

Table 1

Demographic and clinicopathological characteristics of the training, validation cohorts
Variable	Total number, n(%) (n = 7694 )	Training cohort, n(%) (n = 5385 )	Validation cohort, n(%) (n = 2309)
Sex
Male	5433(70.6)	3786(70.3)	1647(71.3)
Female	2261(29.4)	1599(29.7)	662(28.7)
Age(Years)
65–75	4429(57.6)	3102(57.6)	1327(57.5)
≥ 75	3265(42.4)	2283(42.4)	982(42.5)
Race
Black	611(7.9)	442(8.2)	36(1.6)
White	5381(69.9)	3738(69.4)	1175(50.9)
Other	1702(22.1)	1205(22.4)	1098(47.6)
Martial status
Married	7024(91.3)	4906(91.1)	2118(91.7)
Others	670(8.7)	479(8.9)	191(8.3)
Tumor Size (cm)
< 5	4838(62.9)	3366(62.5)	1472(63.8)
≥ 5	2856(37.1)	2019(37.5)	837(27.2)
Grade
I	2541(33.0)	1792(33.3)	749(32.4)
II	3551(46.2)	2465(45.8)	491(21.3)
III	1476(19.2)	1044(19.4)	563(24.4)
IV	126(1.6)	84(1.6)	82(3.6)
T stage
T1	3915(50.9)	2742(50.9)	1173(50.8)
T2	1651(21.5)	1160(21.5)	491(21.3)
T3	1857(24.1)	1294(21.0)	563(24.4)
T4	271(3.5)	189(3.5)	82(3.6)
N stage
N0	7309(95.0)	5113(94.9)	2196(95.1)
N1	385(5.0)	272(5.1)	113(4.9)
M stage
M0	7000(91.0)	4912(91.2)	2088(90.4)
M1	694(9.0)	473(8.8)	221(9.6)
Chemotherapy
Yes	2577(33.5)	1806(33.5)	771(33.4)
No/Unknown	5117(66.5)	3579(66.5)	1538(66.6)
Radiotherapy
Yes	558(7.3)	415(7.7)	143(6.2)
No/Unknown	7136(92.7)	4970(92.3)	2166(93.8)
PVTT
No	7157(93.0)	5015(93.1)	2142(92.8)
Yes	537(7.0)	370(6.9)	167(7.2)
Surgery
No	4253(55.3)	2977(55.3)	1276(55.3)
Yes	3441(44.7)	2408(44.7)	1033(44.7)
Household income($)
< 40000	167(2.2)	131(2.4)	36(1.6)
4000–7000	3896(50.6)	2721(50.5)	1175(50.9)
> 7000	3631(47.2)	22533(47.0)	1098(47.6)

3.2 Selection of prognostic factor

As shown in Table 2,univariate COX analyses based on training cohort shows that Age, Grade, T.stage, N.stage, M.stage, PVTT, Tumor size, Chemotherapy, Radiotherapy, Surgery are associated with OS (P < 0.01).The above 10 variables were included in multivariate Cox analyses, and the results showed: Age, Grade, T.stage, N.stage, M.stage,Chemotherapy, radiotherapy, and surgery were significant to OS (P < 0.01).In addition, we verified the significance of the above eight factors on the overall survival probability of HCC patients using Kaplan-Meier curve and log-rank test (Fig. 2). OS was shorter in patients aged ≥ 75 years than in patients aged 65–75 years(P < 0.01,Fig. 2A);Moderately and well differentiated (GradeI ,II) patients have a better prognosis than poorly differentiated and undifferentiated patients (GradeIII,IV),(P < 0.01,Fig. 2B);Higher TNM staging also led to worse OS.T3,T4 patients have worse prognosis than T1,T2 patients,(P < 0.01,Fig. 2C);OS was significantly worse for patients in N1 and M1 stage than in N0 and M0 stage ,(P < 0.01,Fig. 2D,E);In treatment,Patients who received surgery live significantly longer than those who did not receive surgery,(P < 0.01,Fig. 2F);The median survival time of patients who received chemotherapy was longer than that of patients who did not receive chemotherapy, but the long-term prognosis was worse,(P < 0.01,Fig. 2G);The overall prognosis of patients who received radiotherapy was worse than that of patients who did not receive radiotherapy,(P < 0.01,Fig. 2H).

Table 2

Univariate and multivariate COX analyses of the training cohort on overall survival
Variables	Univariate analysis Multivariate analysis
Variables	HR(95%CI) P-value HR(95%CI) P-value
Sex
Female	Reference
male	1.018(1.015,1.152)	0.016
Age(Years)
65–75	Reference		Reference
≥ 75	1.520(1.434,1.611)	< 0.01*	1.290(1.214,1.370)	< 0.01*
Race
Black	Reference		Reference
White	0.913(0.823,1.014)	0.089	0.958(0.862,1.064)	0.423
Other	0.690(0.614,0.776)	< 0.01*	0.760(0.675,0.856)	< 0.01*
Martial status
Married	Reference
Others	1.112(1.007,1.228)	0.036
Tumor Size
≥ 5	Reference		Reference
༜5	0.800(0.754,0.849)	< 0.01*	1.040(0.976,1.107)	0.229
Grade
I	Reference		Reference
II	0.938(0.879,1.002)	0.056	1.078(1.008,1.152)	0.028*
III	1.315(1.212,1.426)	< 0.01*	1.347(1.239,1.465)	< 0.01*
IV	1.282(1.021,1.609)	0.033	1.493(1.187,1.877)	< 0.01*
T stage
T1	Reference		Reference
T2	1.048(0.972,1.130)	0.219	1.109(1.027,1.197)	0.008*
T3	2.151(2.005,2.309)	< 0.01*	1.598(1.472,1.735)	< 0.01*
T4	2.201(1.890,2.563)	< 0.01*	1.537(1.312,1.801)	< 0.01*
N stage
N0	Reference		Reference
N1	2.673(2.358,3.030)	< 0.01*	1.284(1.121,1.469)	< 0.01*
M stage
M0	Reference		Reference
M1	3.144(2.850,3.469)	< 0.01*	1.826(1.641,2.032)	< 0.01*
PVTT
No	Reference		Reference
Yes	2.147(1.926,2.394)	< 0.01*	1.125(0.996,1.270)	0.058
Chemotherapy
No/Unknown	Reference		Reference
Yes	1.125(1.059,1.195)	< 0.01*	0.647(0.605,0.691)	< 0.01*
Radiotherapy
No/Unknown	Reference		Reference
Yes	1.270(1.144,1.409)	< 0.01*	0.719(0.646,0.801)	< 0.01*
Surgery
No	Reference		Reference
Yes	0.298(0.280,0.317)	< 0.01*	0.283(0.263,0.304)	< 0.01*
Household income($)
< 40000	Reference
4000–7000	0.908(0.753,1.095)	0.313
> 7000	0.805(0.667,0.971)	0.024
*statiscal difference:P < 0.01

3.3 Construction and Validation of the prognostic Nomogram for elderly HCC

For predicting 1-year, 3-year, and 5-year overall survival probabilities, a novel nomogram (Fig. 3) consisting of demographic variables (age), pathologic variables (grading and TNM staging), and treatment modality variables (surgery, radiotherapy, and chemotherapy) was developed using these 8 variables obtained from the Multivariate COX regression in the previous section. The length of the line for each variable in the nomogram represents the contribution of that variable to OS. The nomogram shows that N.stage is the most influential factor to OS, followed by Surgery and M.stage, and T.stage has the least contribution.

As shown in Fig. 4A, in the training cohort, the AUC values for predicting 1-year, 3-year, and 5-year OS in elderly HCC patients were 0.745,0.625, and 0.666, respectively, and in the validation cohort, the AUC values for predicting 1-year, 3-year, and 5-year OS in elderly HCC patients were 0.786,0.648, and 0.677 Fig. 4B. The model has good efficacy in predicting 1-year OS of patients. In addition, the calibration curves showed good agreement between the predicted and observed 1-year, 3-year, and 5-year survival rates of elderly HCC patients in the training cohort, presenting good predictive accuracy (Fig. 4C-E). Similar results were observed in the validation cohort(Fig. 4F-H).

3.4 Risk stratification based on the nomogram

To further validate the significance of the nomogram model in predicting the prognosis of HCC patients.We established a risk classification system by scoring all the patients according to this nomogram and divided the patients in the training set into low-risk group (score < 241) and high-risk group (score ≥ 241) by dichotomization, and plotted the KM curves, which showed that the KM curves showed a significant prognostic difference between the two groups (p < 0.01, Fig. 5A).In the validation cohort, there was also a better prognostic stratification effect(p < 0.01, Fig. 5B).

3.5 Survival outcomes under different treatment modalities after Propensity Score Matching

The effect of different treatment modalities on prognosis was explored by comparing median overall survival between PS-Matched groups. Median survival was significantly higher in the group of patients who underwent surgery (37–42 months) than in no-surgery group(6–7 months)(X² = 1062.980, p < 0.01,Fig. 6A), and the results of all subgroups showed that the surgery group had a better prognosis than the no-surgery group. The median survival (15–18 months) was also significantly higher in the chemotherapy group than in the no-chemotherapy group (6–7 months)( X² = 84.982, p < 0.01,Fig. 6B ), and the results of each subgroup showed that all the prognosis of the chemotherapy group was better than that of the no-chemotherapy group.The median survival (12–15 months) was also significantly higher in the radiotherapy group than in the non-radiotherapy group (6–9 months,X2 = 15.841, p < 0.01,Fig. 6B ), However, in subgroup analysis, it was found that the prognosis of patients who were treated with radiotherapy was poorer than those who did not receive radiotherapy in the T4 subgroup and Grade IV subgroup.

As with many cancers, the prognosis for older patients with HCC is worse than for younger patients as elderly patients are often found to have more comorbidities such as cardiovascular disease, nutritional and metabolic disorders, conditions that can lead to the development of primary cancer(Kendal 2008). At the meantime, due to the lack of effective surveillance, older patients are often diagnosed at an advanced stage(Bakogeorgos, Mountzios et al. 2013, Strøm, Bremnes et al. 2015, Sun, Liu et al. 2022). Coupled with the current international environment of an aging population, this article hence focuses on the prognostic factors affecting hepatocellular carcinoma in the elderly.The main method for assessing HCC disease progression, treatment options, and prognosis is the TNM staging system released by the American Joint Committee on Cancer (AJCC), but this staging system contains fewer clinical parameters and is applied to all cancer populations, which is not targeted enough to predict the prognosis of HCC, especially in elderly patients with HCC. Nomogram based on AJCC staging combined with other clinical parameters, demographic characteristics have been widely used as a convenient and effective tool for prediction of survival time, and their accuracy and validity have been validated in other cancers (Mao, Fu et al. 2021, Tong, Li et al. 2021, Wang, Zhanghuang et al. 2022).In this article, we successfully attempted to predict OS in elderly patients with hepatocellular carcinoma using the construction of a combined nomogram with the above parameters, in combination with clinical parameters(Pathologic grading and TNM staging), demographic characteristics(Age), and treatment information(surgery, radiation and chemotherapy).OS decreases significantly with the increase of age. Therefore, more attention should be paid to elderly patients with hepatocellular carcinoma in clinical.However, unlike the findings of Xu(Xu, Qin et al. 2023), in this paper, PVTT was not statistically significant to OS. Our results suggest that this nomogram has satisfactory accuracy and discriminative ability in predicting 1-year OS, but poorer efficacy in predicting 3- and 5-year OS, which may be related to the fact that our study population was an elderly group, whose own characteristics determine that their long-term survival is affected by numerous factors, all of which may contribute to the decline in the predictive efficacy of the nomogram.In addition, we used calibration curves in the training and validation cohorts to verify whether there is a good agreement between the predicted and actual observed survival rates of nomogram, and the results presented good prediction accuracy.Next, we introduce a scoring system based on nomograms, which calculates the contribution of each variable in the nomogram to OS according to the equation to obtain a point, and the score of each patient is equal to the sum of the points of these variables.Patients were then divided into high-risk and low-risk groups based on the scores, and this distinction was confirmed in the training cohort and validation cohort to be significant for OS.With the assistance of this risk stratification system methodology, risk stratification of patients can be achieved in the clinic to increase the attention to the high-risk group and to improve the poor prognosis of these patients by taking active interventions as early as possible.

We also considered the drawbacks of using the SEER database and remedied them:(1) (1)Because the database collects information on cancer patients in the United States, the variable of race is somewhat biased, so race information was not used to create the regression equation.(2)Since the treatment of HCC patients is determined by the physician based on the patient's condition, i.e., the decision for a patient to undergo surgery/radiation/chemotherapy is the result of the physician's careful patient selection.This non-randomization bias may affect the comparison of overall survival between patients, this limitation can only be tackled by the RCT approach, but based on ethical considerations, we cannot randomize patients’ treatment modalities in order to carry out a research without taking into account their specific conditions.To minimize data bias and interference by confounding factors, we grouped patients with surgery, chemotherapy, and radiotherapy separately, and then used PSM to match other variables, fully considering the independent correlations of all other variables with the grouped variables, to further compare the differences in patient survival between treatment modalities more accurately,which can help in the selection of optimal clinical treatment decisions for patients.

In this part of the results we found that (1) Dividing the patients into radiotherapy and non-radiotherapy groups, PSM controlled for bias and the results showed that although the median overall survival was significantly higher in the radiotherapy group (12–15 months) than in the non-radiotherapy group (6–9 months, Fig. 6C).However, when subgroups were analyzed, it was found that in the T4 subgroup, radiotherapy patients had a worse prognosis than those who did not receive radiotherapy, and the same results were found in the Grade IV subgroup.This is different from the findings of Xu et al(Xu, Qin et al. 2023). In response to this result, we have two considerations .Firstly, it may suggest that in elderly patients with poorly staged hepatocellular carcinoma, radiotherapy causes destruction of surrounding normal tissues and may not be conducive to long-term prognosis, especially in the elderly group(Seong 2009).Moreover, in recent years, many studies have shown that many cancer patients end up dying not from the primary cancer but from complications after radiotherapy, the most common complications being severe infections secondary to immunocompromised, radioactive ulcers(Seong 2009, Peixoto, Hammond et al. 2014).The second is that this study did not continue to group the patients according to the radiotherapy method, and the effect of different types of radiotherapy on the prognosis may be different. In recent years, many studies have confirmed that stereotactic radiotherapy, compared with conventional radiotherapy, permits a higher dose of treatment to the tumor site, and plays a more favorable role in the long-term prognosis of patients with hepatocellular carcinoma than conventional radiotherapy(Ohri, Dawson et al. 2016). The specific radiotherapy modality of the patients in the radiotherapy group in our paper is not clear, which may also cause some bias to the results.(2) As with other findings,among the surgery and no-surgery groups, we found that the overall survival of patients in the surgery group was significantly longer than that of no-surgery patients(Hu, You et al. 2019, He, Chen et al. 2021, Su, Shen et al. 2023). Subgroup analyses also showed that surgery was not only beneficial for patients aged 65–75 years, with well-differentiated tumors, and who had been treated with radiotherapy, but also had a positive impact on patients with poor prognostic factors (including age ≥ 75 years and poorly-differentiated tumors), and could improve the survival of this group of patients; therefore, it is worthwhile to further explore the possibility of expanding the indications for surgery in a follow-up study.Although this article develops the first nomogram on OS prediction in elderly HCC patients, its limitations should be noted. Firstly, this paper has not been externally validated. Further external validation using a multicenter cohort with a large sample size is necessary to confirm the performance of nomogram. Secondly, we did not specific analyze the effect of different surgical modalities and different radiotherapy regimens on the prognosis of HCC, which still has a certain degree of bias. Finally, the SEER database lacks some other prognostic factors, such as surgical margin status, laboratory data (bilirubin, albumin, AFP, etc.), which are not included in the nomogram and may also be a certain degree of decline in the nomogram efficacy, all of which should be evaluated in future studies.Conclusion

The nomogram constructed in this article based on demographic characteristics, clinical parameters and treatment information effectively fills the gap in prognosis prediction of elderly HCC patients, and the risk stratification system can effectively identify high-risk patients for early intervention to improve their poor prognosis. Therefore, this nomogram can be used as an effective tool to guide the individualized treatment and follow-up management of elderly HCC patients.

Competing interest: No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

# Shuyu Liu and Mengyue Zhou contributed equally to this study.

Ethics approval and consent to participate: Since the patient data extracted from SEER database were de-identified and available to the public for research purposes, informed patient consent was not required for this study and ethical approval was waived by the local ethics committee.

Consent for publication:Not applicable.

Availability of data and materials:The data analyzed in this study were obtained from the Surveillance, Epidemiology, and End Results (SEER) database. The following licenses/restrictions apply: requests to access these database must first be approved by SEER. Requests to access these datasets should be directed at https://seer.cancer.gov/data/ access.html.

Competing interest: No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

# Shuyu Liu and Mengyue Zhou contributed equally to this study.

Funding: None

Authors’ contributions: Shuyu Liu. and Lin Miao conceived the idea for the review. Shuyu Liu and Mengyue Zhou. performed the literature search, data extraction and analysis. Shuyu Liu. and Jiayi Gu ,,Peng Chen,Xueting Deng drafted the article. All authors have read the manuscript and given their approval for publication.

Acknowledgements:First and foremost, I appreciate my country who give me a comfortable learning atmosphere. Second, I would like to show my deepest gratitude to my supervisor, Prof. Linmiao, who has walked me through all the stages of the writing of this thesis.

Bakogeorgos M, Mountzios G, Kotsantis G, Economopoulou P, Fytrakis N, Kentepozidis N. Chemotherapy compliance, tolerance and efficacy in elderly and non-elderly patients with metastatic colorectal cancer: a single institution comparative study. " J buon. 2013;18(3):629–34.
Bruix J, Gores GJ, Mazzaferro V. Hepatocellular carcinoma: clinical frontiers and perspectives. Gut. 2014;63(5):844–55.
Cai W, Fan J, Shen T, Yu J. The Influence of Marital Status on the Survival of Patients with Uveal Melanoma. J Ophthalmol. 2020;2020:7012940.
Chen Z, Yin K, Zheng D, Gu J, Luo J, Wang S, Chen H. Marital status independently predicts non-small cell lung cancer survival: a propensity-adjusted SEER database analysis. J Cancer Res Clin Oncol. 2020;146(1):67–74.
Fan J, Wu ZQ, Tang ZY, Zhou J, Qiu SJ, Ma ZC, Zhou XD, Ye SL. Multimodality treatment in hepatocellular carcinoma patients with tumor thrombi in portal vein. World J Gastroenterol. 2001;7(1):28–32.
He T, Chen T, Liu X, Zhang B, Yue S, Cao J, Zhang G. A Web-Based Prediction Model for Cancer-Specific Survival of Elderly Patients With Early Hepatocellular Carcinoma: A Study Based on SEER Database. Front Public Health. 2021;9:789026.
Hu Y, You S, Yang Z, Cheng S. Nomogram predicting survival of hepatocellular carcinoma with portal vein tumour thrombus after curative resection. ANZ J Surg. 2019;89(1–2):E20–e25.
Kendal W. Dying with cancer: the influence of age, comorbidity, and cancer site. Cancer. 2008;112(6):1354–62.
Li N, Feng S, Xue J, Wei X, Shi J, Guo W, Lau W, Wu M, Cheng S, Meng Y. Hepatocellular carcinoma with main portal vein tumor thrombus: a comparative study comparing hepatectomy with or without neoadjuvant radiotherapy. HPB: the official journal of the International Hepato Pancreato Biliary Association. 2016;18(6):549–56.
Mao W, Fu Z, Wang K, Wu J, Xu B, Chen M. Prognostic nomogram for patients with lung metastatic renal cell carcinoma: a SEER-based study. Ann Palliat Med. 2021;10(3):2791–804.
Ohri N, Dawson L, Krishnan S, Seong J, Cheng J, Sarin S, Kinkhabwala M, Ahmed M, Vikram B, Coleman C, Guha C. (2016). Radiotherapy for Hepatocellular Carcinoma: New Indications and Directions for Future Study. J Natl Cancer Inst 108(9).
Peixoto D, Hammond S, Issa N, Madan R, Gill R, Milner D, Colson Y, Koo S, Baden L, Marty F. Green herring syndrome: bacterial infection in patients with mucormycosis cavitary lung disease. Open forum infectious diseases. 2014;1(1):ofu014.
Seong J. Challenge and hope in radiotherapy of hepatocellular carcinoma. Yonsei Med J. 2009;50(5):601–12.
Strøm HH, Bremnes RM, Sundstrøm SH, Helbekkmo N, Aasebø U. How Do Elderly Poor Prognosis Patients Tolerate Palliative Concurrent Chemoradiotherapy for Locally Advanced Non-Small-Cell Lung Cancer Stage III? A Subset Analysis From a Clinical Phase III Trial. Clin Lung Cancer. 2015;16(3):183–92.
Su K, Gu T, Xu K, Wang J, Liao H, Li X, Wen L, Song Y, Zhong J, He B, Liu X, He J, Liu Y, Li Q, Feng X, Chen S, Yang B, Huang W, Jin H, Luo X, Hu T, Chen J, Wu Z, Lu S, Zhang J, Rao M, Xie Y, Wang J, Zhu X, Chen L, Li B, Su S, Yang X, Wang J, Zeng H, Wang P, Yan M, Chen X, He K, Han Y. Gamma knife radiosurgery versus transcatheter arterial chemoembolization for hepatocellular carcinoma with portal vein tumor thrombus: a propensity score matching study. Hepatol Int. 2022;16(4):858–67.
Su K, Shen Q, Tong J, Gu T, Xu K, Li H, Chi H, Liu Y, Li X, Wen L, Song Y, Guo Q, Chen J, Wu Z, Jiang Y, He K, Guo L, Han Y. Construction and validation of a nomogram for HBV-related hepatocellular carcinoma: A large, multicenter study. " Ann Hepatol. 2023;28(4):101109.
Sun H, Liu M, Yang X, Ren Y, Dai H, Wang C. Construction and validation of prognostic nomograms for elderly patients with metastatic non-small cell lung cancer. Clin Respir J. 2022;16(5):380–93.
Sung H, Ferlay J, Siegel R, Laversanne M, Soerjomataram I, Jemal A, Bray F. (2021). "Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries." CA: a cancer journal for clinicians 71(3): 209–249.
Tong Y, Li J, Huang Y, Zhou J, Liu T, Guo Y, Yu J, Zhou S, Wang Y, Chang C. Ultrasound-Based Radiomic Nomogram for Predicting Lateral Cervical Lymph Node Metastasis in Papillary Thyroid Carcinoma. " Acad Radiol. 2021;28(12):1675–84.
Tortora R, Farella N, Morisco F, Coppola C, Izzo F, Salomone Megna A, Federico A, Messina V, Nardone G, Piai G, Ragone E, Adinolfi L, D'Adamo G, Stanzione M, Francica G, Torre P, De Girolamo V, Coppola N, Guarino M, Dallio M, Rocco L, Di Costanzo G. Development of a risk score to predict portal vein tumor thrombosis in patients with hepatocellular carcinoma. Eur J Gastroenterol Hepatol. 2023;35(7):734–41.
Wang J, Zhanghuang C, Tan X, Mi T, Liu J, Jin L, Li M, Zhang Z, He D. A Nomogram for Predicting Cancer-Specific Survival of Osteosarcoma and Ewing's Sarcoma in Children: A SEER Database Analysis. Front Public Health. 2022;10:837506.
Wei X, Jiang Y, Zhang X, Feng S, Zhou B, Ye X, Xing H, Xu Y, Shi J, Guo W, Zhou D, Zhang H, Sun H, Huang C, Lu C, Zheng Y, Meng Y, Huang B, Cong W, Lau WY, Cheng S. Neoadjuvant Three-Dimensional Conformal Radiotherapy for Resectable Hepatocellular Carcinoma With Portal Vein Tumor Thrombus: A Randomized, Open-Label, Multicenter Controlled Study. J Clin Oncol. 2019;37(24):2141–51.
Weir HK, Thompson TD, Soman A, Møller B, Leadbetter S. The past, present, and future of cancer incidence in the United States: 1975 through 2020. Cancer. 2015;121(11):1827–37.
Xu C, Liu X, Chen Y, Mao Y, Guo R, Zhou G, Tang L, Lin A, Sun Y, Ma J. Impact of marital status at diagnosis on survival and its change over time between 1973 and 2012 in patients with nasopharyngeal carcinoma: a propensity score-matched analysis. Cancer Med. 2017;6(12):3040–51.
Xu J, Qin S, Wei H, Chen Y, Peng Y, Qi L. Prognostic factors and an innovative nomogram model for patients with hepatocellular carcinoma treated with postoperative adjuvant transarterial chemoembolization. Ann Med. 2023;55(1):2199219.

No competing interests reported.

Download PDF

Reviews received at journal
16 Jul, 2024
Reviewers agreed at journal
29 May, 2024
Reviewers invited by journal
03 Jan, 2024
Editor assigned by journal
02 Jan, 2024
Editor invited by journal
11 Oct, 2023
Submission checks completed at journal
11 Oct, 2023
First submitted to journal
28 Sep, 2023

You are reading this latest preprint version

A nomogram for the prediction of overall survival and Prognostic significance of different treatment modalities for elderly Hepatocellular carcinoma

Status:

Version 1

Abstract

Figures

1. Introduction

2. Materials and method

2.1 Patients cohort

2.2 Statistical analysis

2.3 Propensity Score-Matching

3. Results

3.1 Demographic and clinical characteristics

3.2 Selection of prognostic factor

3.3 Construction and Validation of the prognostic Nomogram for elderly HCC

3.4 Risk stratification based on the nomogram

3.5 Survival outcomes under different treatment modalities after Propensity Score Matching

Discussion

Declarations

References

Additional Declarations

Status:

Version 1