Cancer diagnosis and prognosis after initiation of hemodialysis: multicenter Japan CANcer and DialYsis (J-CANDY) study.

Текст сохранён. Часть иллюстраций пока доступна только в первоисточнике; загрузка будет проверена повторно.

What was known:

Cancer is the leading cause of death among hemodialysis patients; therefore, the management of cancer in dialysis patients is increasingly important to improve their prognosis.

Surgical therapy has been the cornerstone of treating patients with solid cancers, however little is known about the clinical practice of surgical therapy and long-term postoperative outcomes of hemodialysis cancer patients.

Postoperative outcomes and factors affecting mortality in hemodialysis cancer patients who have undergone surgery remain unclear.

This study adds:

The primary cause of death among patients who underwent surgery was non-cancer-related death, whereas that of patients who did not undergo surgery was cancer-related death.

Pancreatic cancer and anemia were poor prognostic factors for patients who underwent surgery.

Surgery rate and 3-year overall survival of hemodialysis cancer patients might be equivalent to that of general cancer patients.

Potential impact:

This study helps predict postoperative prognosis in hemodialysis cancer patients.

This study suggests that surgery is a treatment option for hemodialysis cancer patients if dialysis patients are diagnosed with resectable cancer.

INTRODUCTION

Previous studies have suggested that patients on hemodialysis have a higher incidence [ 1 ] and mortality [ 2 , 3 ] of cancer than the general population. In Japan, cancer is the third leading cause of death among hemodialysis patients, and the crude ratio per all-cause death has been gradually increasing [ 4 ]. Therefore, management of cancer in patients on dialysis is becoming increasingly important to improve their outcomes. However, data are lacking on the clinical practice to manage cancer in patients on hemodialysis, and the disease status at diagnosis, the first-line therapy selected and the factors affecting survival. The CANcer and DialYsis (CANDY) study is a major study on hemodialysis cancer management in France, which mainly focuses on types of chemotherapy and dose adjustment [ 5 ], but not on surgical treatments.

Surgeries have been the cornerstone of the treatment of non-hemodialysis patients with solid cancers [ 6 ]. While Miyamoto et al . reported that hemodialysis cancer patients have higher rates of postoperative comparisons and 30-day mortality [ 7 ], there are limited data available on long-term postoperative outcomes comparing cancer patients on hemodialysis with cancer patients in general [ 7–10 ]. Therefore, it remains unclear whether surgical treatment improves the prognosis of hemodialysis cancer patients. Additionally, factors affecting mortality in hemodialysis cancer patients remain unclear.

We conducted a nationwide cohort study of patients on hemodialysis with cancer in Japan, using an available longitudinal dataset. The primary objective of our study was to assess the long-term prognosis of patients on hemodialysis with cancer compared with general cancer patients. Furthermore, we sought to define the clinical characteristics and prognostic differences between patients with cancer on hemodialysis who had undergone surgery and those who did not, and to uncover the clinical practice of hemodialysis cancer management in Japan, the leading dialysis country in Asia.

MATERIALS AND METHODS

The Onconephrology Consortium in Japan

This retrospective cohort study was conducted by the Onconephrology Consortium in Japan with clinical investigators, both medical oncologists and nephrologists, from 22 high-volume cancer centers, including 8 university hospitals, with a mean of 768 beds and 2186 cancer patients per year ( Supplementary data, Table S1 ). Data of dialysis cancer patients were collected from 20 of those facilities.

Study design and population

A total of 674 patients who were undergoing chronic hemodialysis and were newly diagnosed with cancer from January 2010 to December 2012 in the participating institutes ( Supplementary data, Table S1 ) were assessed (Fig. 1 ). Following screening, 502 patients were included in the study. The observation period was from the time of cancer diagnosis until October 2015. This study protocol was approved by the ethics committees of Kyoto University Graduate School and Faculty of Medicine (R2356) [ 11 ].

Exclusion criteria

The exclusion criteria were as follows: patients with less common cancers (<20 patients each) and patients with missing data regarding primary therapy. According to the exclusion criteria, patients whose primary cancer sites were oropharynx ( n = 14), larynx ( n = 7), esophagus ( n = 19), gall bladder/bile duct ( n = 6), skin ( n = 16), uterus ( n = 14), ovary ( n = 10), prostate ( n = 13), urothelium ( n = 4), central nervous system ( n = 2), thyroid gland ( n = 15), malignant lymphoma ( n = 11), multiple myeloma ( n = 6), leukemia ( n = 4) and miscellaneous ( n = 26) were excluded. Patients with missing data regarding primary therapy were also excluded ( n = 5).

Data collection

The following demographic and clinical data for each participant were obtained (Table 1 ): age; sex; indication for hemodialysis, including diabetes (DM), chronic glomerulonephritis (CGN), hypertension (HT) and other; dates of hemodialysis initiation and cancer diagnosis; symptoms at cancer diagnosis (symptomatic or asymptomatic); methods for cancer diagnosis (routine screening test or other); primary cancer therapy; degree of cancer progression; and hemoglobin (Hb) and serum albumin (Alb) levels at cancer diagnosis. Follow-up was performed by review of medical records. In addition, data on all-cause death or censoring and cause of death based on cancer relevance were collected at the end of follow-up due to death or censoring. Missing data on event dates were imputed as follows: for missing dates (with available months), mid-month (the 15th) was substituted ( n = 37 and 14, respectively), and for missing months, mid-year date (June 15) was substituted ( n = 5 and 0, respectively). Sensitivity analysis was performed after excluding patients with missing hemodialysis initiation and cancer diagnosis dates ( Supplementary data, Table S2 ). In line with previous reports, the responsible physician at each facility classified the degree of initial cancer condition at diagnosis as follows: resectable disease, locally advanced disease or metastatic disease [ 11 ]. Primary therapy was categorized into surgery including endoscopic resection; chemotherapy; best supportive care (BSC); or other. The study endpoint was all-cause mortality. An adequate Hb level was defined as Hb 10–12 g/dL [ 12 ]. Because hypoalbuminemia was defined as Alb <3.5, <3.0 or 2.5 g/dL in previous reports regarding hypoalbuminemia and mortality [ 13 ], we defined hypoalbuminemia as Alb <3.0 g/dL in this study. Informed consent was obtained using an opt-out process on the website. We followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines to report this study [ 14 ].

Statistical analyses

The results of the continuous variables are expressed as means ± standard deviations (SD) (age, Hb and Alb level) or medians and interquartile ranges (time from hemodialysis to cancer diagnosis). Statistical analyses were performed using JMP (version 11; SAS Inc., Cary, NC, USA), GraphPad Prism for Mac (version 9.50; GraphPad Software, La Jolla, CA, USA) and R (version 4.0.2; https://www.R-project.org/ ). Multivariate analyses were conducted using the Cox proportional hazards model. We chose gastrointestinal and liver cancers as the reference for hazard ratio analysis because these categories had the largest numbers of patients. Among the total cohort of 502 patients, we could follow prognosis in 500 patients. Among the 500 patients, 229 patients were censored before 36 months after cancer diagnosis, and we analyzed outcomes during a 3-year follow-up, but not longer. We performed a multicenter study to reduce selection bias and multivariate analyses to prevent confounding in the analysis of prognostic factors. Additionally, patients who underwent surgery and those who did not receive similar examinations and treatments from the start of dialysis until the cancer diagnosis.

RESULTS

Clinical characteristics of patients

Of the 674 patients enrolled, 502 (74%) were diagnosed with cancers of interest (kidney, colorectum, stomach, lung, liver, bladder, breast and pancreas) (Fig. 1 ), which are prevalent in hemodialysis patients [ 5 , 15 , 16 ]. The baseline characteristics of the patients are summarized in Table 1 . Age at cancer diagnosis was 67.9 ± 10.4 years (mean ± SD), and 73% of the patients were male. Underlying causes of end-stage kidney disease included DM (32%), CGN (32%) and HT (9%), similar to those of general hemodialysis patients in Japan [ 17–19 ]. Kidney cancer was the most common (32%), followed by cancers in colorectum (17%) and stomach (14%).

Time from hemodialysis initiation to cancer diagnosis, presence of symptoms at diagnosis and reason for diagnosis

The median time from hemodialysis initiation to cancer diagnosis was 74 months (Table 1 ). Among cancer types, the median time from hemodialysis initiation to diagnosis of kidney and breast cancer was longer: 142 and 156 months, respectively, compared with those of bladder, liver and stomach cancer, 31, 39, and 40 months, respectively.

In total, 206 patients (41%) were symptomatic and 287 (57%) were asymptomatic at diagnosis (Table 1 ). Of the 287 asymptomatic patients, 217 (76%) were incidentally diagnosed by routine screening at hemodialysis.

Clinical characteristic of patients stratified by surgery implementation

Of our cohort of 502 patients, 370 (74%) patients underwent surgery (Fig. 1 ). We next compared various clinical factors between patients who underwent surgery (surgery group) and those who did not (non-surgery group) (Table 1 ). In the surgery group, kidney cancer was the most common (36%), followed by colorectal (18%) and stomach (17%) cancers. In the non-surgery group, lung cancer was the most common (29%), followed by liver (20%) and kidney (19%) cancers. Dialysis vintage to cancer diagnosis was 78 months in the surgery group and 64 months in the non-surgery group. In contrast, dialysis vintage to cancer diagnosis of patients with kidney cancer was 140 months in the surgery group and 221 months in the non-surgery group. Additionally, levels of Hb and serum Alb, which are prognostic determinants in cancer patients [ 20 ], were comparable between the two groups. In the surgery group, 30 patients received perioperative chemotherapy, and 27 (90%) patients among them received postoperative chemotherapy. The remaining 340 patients received surgical therapy only. We assessed primary therapy in the non-surgery group (Table 1 , Fig. 2 ). Among 132 patients in the non-surgery group, 44 (33%) patients underwent chemotherapy, 42 (32%) were offered BSC and 46 (35%) underwent other therapy (Table 1 ). Most patients with kidney, colorectum, stomach and pancreas cancers selected chemotherapy or BSC, whereas 17 (45%) patients with lung cancer received radiation and 16 (59%) with liver cancer received hepatic arterial infusion of chemotherapy (Fig. 2 A). Hepatic arterial infusion of chemotherapy was used only for primary liver cancer. Among patients with resectable who did not undergo surgery, 7 (88%) with lung cancer received radiation and 10 (67%) with liver cancer received hepatic arterial infusion of chemotherapy (Fig. 2 B). Among patients with locally advanced or metastatic disease, 11 (37%) with lung cancer received radiation and 6 (50%) with liver cancer received hepatic arterial infusion of chemotherapy (Fig. 2 C).

Initial cancer conditions when diagnosed

Initial cancer condition at the time of diagnosis was categorized as resectable, locally advanced or metastatic (Table 1 ). Among the 370 patients in the surgery group, 365 (99%) were evaluated as having resectable cancer. In contrast, among 132 patients in the non-surgery group, 30 (23%) had resectable cancers, 34 (26%) had locally advanced cancers, and 68 (52%) had metastatic cancers.

Next, we evaluated initial cancer conditions in each group (Fig. 3 ). In total, more than 80% of the patients had resectable kidney, colorectal, stomach, bladder and breast cancers. In contrast, approximately half of the patients with lung, liver and pancreatic cancers had locally advanced or metastatic disease (Fig. 3 A). In the surgery group, almost all patients were evaluated as having resectable cancer (Fig. 3 B). However, in the non-surgery group, most patients with kidney, colorectal and stomach cancers had metastatic disease (Fig. 3 C). Eight (21%) patients with lung cancer were evaluated as having resectable cancer but did not undergo surgery (Fig. 3 C). Further, we analyzed cancer stages in each group and found that among the patients not undergoing surgery, 69% of patients with liver cancer and 36% of patients with lung cancer were in stage I or II ( Supplementary data, Fig. S1 ). This trend was similar to that seen when classifying patients by initial cancer condition (Fig. 3 ).

We also evaluated symptoms at diagnosis ( Supplementary data, Table S3 ). Among patients with resectable disease, 241 (61%) were asymptomatic at diagnosis. Of these asymptomatic patients, 184 (76% of those who were asymptomatic at diagnosis) were incidentally diagnosed by routine screening tests at hemodialysis. In contrast, among patients with metastatic disease, 43 (61%) were symptomatic at diagnosis. We then compared cancer progression between our study patients and the general patients from the National Cancer Center (NCC) database [ 21 ], which included 339 376 patients diagnosed with cancer at 349 facilities in Japan in 2012. Data for cancer stages were missing in 144 (29%) patients in J-CANDY. To compare the outcomes using data from two separate registries, we used the following interpretation to allow comparison of cancer progression stages. In general, stage I and II cancers are resectable, and stage IV cancers are mostly unresectable; stage III cancers may or may not be resectable, with some exceptions [ 22 ]. We compared the percentages of patients with resectable disease in our cohort with cancer progression stages of the general patients ( Supplementary data, Table S4 ) [ 21 ]. For kidney cancer, patients with resectable disease were 84% in our study, whereas general patients with stage I or II were 71% and those with stage I, II or III were 83%. For stomach cancer, patients with resectable disease were 92% in our study, whereas patients with stage I or II were 70% and those with stage I, II or III were 80%.

Mortality of patients on hemodialysis with cancer

One hundred and twenty-nine (26%) patients died during the entire follow-up, and the median follow-up was 26 months from cancer diagnosis. 117 (23%) patients died during the 3-year follow-up (Table 2 ). After excluding missing data, the median follow-up length and number of patients who died during the 3-year follow-up period were similar, 26 months and 103 (23%), respectively. The 3-year overall survival of patients in the present study was 70%, which was comparable to that of general hemodialysis patients (73%) [ 23 ].

We analyzed overall survival in the surgery and non-surgery groups. Three-year overall survival was 83% and 32% in patients in the surgery and non-surgery groups, respectively (Table 2 ). In the surgery group, patients with kidney and breast cancer had 3-year survival rates exceeding 90%, whereas those with pancreatic cancer had 3-year survival rates of approximately 40%. In contrast, in the non-surgery group, while patients with liver cancer had a 3-year survival rate of 64%, those with stomach or pancreatic cancer had a 3-year survival rate of 0%. The causes of death were different in the surgery group compared with the non-surgery group. Non-cancer-related death (80%) was most common in the surgery group, whereas cancer-related death (70%) was predominant in the non-surgery group.

Multivariate analysis of factors associated with overall survival in hemodialysis cancer patients with or without surgery

We next aimed to identify factors associated with mortality in the present study, according to surgery implementation (Table 3 ). Pancreatic cancer and anemia were associated with high mortality rates in the surgery group. Pancreatic cancer was also associated with a high mortality rate in the non-surgery group.

Comparison of surgery rate between hemodialysis patients and non-hemodialysis patients with cancer

In this study, 74% of hemodialysis cancer patients underwent surgery as the first-line therapy (Table 1 ). We compared their surgery rates with those of patients enrolled in the NCC (Fig. 4 ) [ 21 ]. Surgery rates of hemodialysis patients and the general patients were 84% and 84% for kidney cancer, 82% and 87% for colorectal cancer, 23% and 34% for liver cancer, and 44% and 28% for pancreatic cancer, respectively.

Comparison of 3-year survival ratio between hemodialysis patients and the general patients

We further compared cancer prognosis between hemodialysis cancer patients and the general patients enrolled in the NCC (Fig. 5 ) [ 21 ]. The 3-year survival rates of hemodialysis patients and the general patients were 84% and 81% for kidney cancer, 76% and 73% for colorectal cancer, and 67% and 50% for liver cancer, respectively (Fig. 5 A).

Next, we analyzed 3-year survival rates of the patients who underwent surgery. The 3-year survival rates of hemodialysis patients and the general patients were 93% and 90% for kidney cancer, 88% and 81% for colorectal cancer, and 73% and 83% for stomach cancer, respectively (Fig. 5 B).

The 3-year survival rates among patients who did not undergo surgery were lower in both groups than in patients in the surgery group (Fig. 5 C).

DISCUSSION

This multicenter study reports clinical practices of cancer diagnosis, treatment and prognosis in hemodialysis patients in Japan stratified by the implementation of surgery. In the surgery group, pancreatic cancer and anemia were associated with high mortality rates. The likelihood of undergoing surgery for cancer in hemodialysis patients was comparable to that in general patients. Furthermore, the survival of hemodialysis patients with cancer was similar to that of general patients.

Our cohort study revealed a reasonable 3-year survival outcome in patients on hemodialysis with cancers undergoing surgeries. Previous nationwide, population-based cohort studies of hemodialysis cancer patients failed to determine the survival outcomes of surgical intervention [ 15 , 16 , 24–26 ], and a small cohort studies of dialysis patients undergoing surgery for cancers have not analyzed patients stratified by surgery implementation [ 27–31 ]. Our study also demonstrated that the major cause of death in the surgery group was not cancer-related death (Table 2 ), attributed to infection, heart failure and cerebrovascular accident (Table 2 ), all of which were the major causes of mortality in hemodialysis patients in the USA and Japan [ 32 , 33 ], as hemodialysis cancer patients in the surgery group tended to have less advanced cancer, and surgeries could be curative. In contrast, hemodialysis cancer patients in the non-surgery group were more likely to die from cancer because they tended to have more advanced cancer. These results are essential for estimating the prognosis of hemodialysis cancer patients, and they suggest that the treatment of hemodialysis comorbidities is important in improving the survival of hemodialysis cancer patients with cancer.

While our cohort study was conducted in Japan, cancer outcomes in dialysis patients are of interest in other countries as well. In France, the CANDY study focused on the management of hemodialysis cancer patients [ 5 ]. Interestingly, the time to cancer diagnosis, proportion of surgical patients and overall survival are quite different from those of the current study. In our study, the median time from hemodialysis initiation to cancer diagnosis was 74 months (Table 1 ), which was much longer than that of the CANDY study (13 months). The 2-year overall survival of our patients was 76% (Table 2 ), which was higher than that of the CANDY study (42%). This difference in survival may be explained by the several factors related to different clinical characteristics of dialysis patients between in Western countries and in Japan. Firstly, 74% of our patients underwent surgery (Table 1 ), which was much higher than the percentage in the CANDY study (37%). In addition, 9% of our patients underwent chemotherapy (Table 1 ), which was lower than the percentage reported in the CANDY study (28%). These findings may imply that more patients in Japan than in France were diagnosed with cancer in a resectable stage, which might have led to improved survival in this study. Secondly, the choice of modalities of renal replacement therapies varies considerably between countries. The number of kidney transplants per million inhabitants is 73.6 in the USA, 54 in France and 13.8 in Japan [ 34–36 ], whereas the number of dialysis patients is 1546 in the USA, 745 in France and 2640 in Japan [ 35 , 37 , 38 ]. Because of the limited availability of kidney transplants in Japan, even relatively healthy patients who would have been eligible for kidney transplantation in the USA or Europe often remain on dialysis. Furthermore, differences in hemodialysis practice patterns and prevalence of comorbidities might contribute to the survival difference [ 39 ]. The 1-year mortality rate for dialysis patients is 6.6% in Japan, compared with 21.7% in the USA and 15.6% in Europe [ 39 ]. These overall low mortality rates in dialysis patients may hold true for the subgroup of dialysis patients with cancers. Our study may help determine the prognosis of hemodialysis cancer patients in countries with higher prevalence of hemodialysis. Our study may help determine the prognosis of hemodialysis cancer patients in countries with higher prevalence of hemodialysis.

Our study has potential implications on cancer screening strategies for patients on hemodialysis. We found that kidney and breast cancer were associated with longer dialysis vintage (Table 1 ), consistent with a previous report [ 40 , 41 ]. A study from Taiwan reported an average dialysis vintage at the time of breast cancer diagnosis is 40.6 months with a 10-year follow-up [ 16 ], vs 156 months in our study with up to 30-year follow-up. The results of our study suggests that breast cancer should be considered in patients with long dialysis vintage. We also analyzed the symptoms and triggers of detection at cancer diagnosis. In our study, 57% patients were diagnosed asymptomatically, of which 76% were detected by routine screening tests (Table 1 ), which accounts for 61% and 76% of patients with resectable cancers, respectively. ( Supplementary data, Table S3 ). Additionally, our results suggest that patients receiving dialysis in our cohort are diagnosed with cancers in earlier stages compared with general patients ( Supplementary data, Table S4 ). In Japan, dialysis patients are recommended to receive routine cancer screening tests, including fecal occult blood test, gastrointestinal endoscopy and abdominal ultrasound [ 42 ]. However, we should be cautious about this conclusion because lead time bias, length bias and overdiagnosis could be factors [ 43 ]. In addition, the efficacy of screening tests for dialysis patients is controversial. For dialysis patients with limited life expectancy, screening tests are considered to be cost insufficient and not recommended [ 44 , 45 ], however dialysis patients who are expected to have a longer life expectancy may benefit from cancer screening tests [ 46 ]. Future studies are needed to determine the subgroup of dialysis patients who may benefit from cancer screening tests. Further, we might have overestimated the ratio of asymptomatic patients because most dialysis patients are symptomatic, such as fatigue and pain [ 47 ], which might conceal cancer symptoms.

Pancreatic cancer and anemia were associated with high mortality in the surgery group (Table 3 ), which was consistent with the results of previous studies in the general patients [ 48 ]. These results may help predict postoperative prognosis in hemodialysis cancer patients.

More importantly, comparison with NCC data revealed that surgery rates for dialysis patients and the general patients were comparable, with the exception of lower rates for dialysis patients with liver cancer and higher rates for dialysis patients with pancreatic cancer (Fig. 4 ). The 3-year survival rates for all patients were also comparable, except for higher rates for liver cancer in dialysis patients (Fig. 5 ). Additionally, the 3-year survival rates for dialysis patients and the general patients who underwent surgery were comparable, except for stomach cancer. These results suggest that surgical therapy effectively treats cancer in dialysis patients. A previous single-center study showed that dialysis patients have a similar prognosis to non-dialysis patients [ 27 ], which is consistent with our results. In our study, the 3-year survival rate for hemodialysis patients with cancer was 70%, consistent with the 3-year survival rate (73%) for general hemodialysis patients [ 23 ]. This may be because the cancer diagnosis in approximately half of the patients in our study was made by screening, leading to better prognosis [ 49 ]. Among all patients who did not undergo surgery, those with kidney and colorectal cancers showed similar survival rates between hemodialysis patients and general patients, whereas the prognosis for liver cancer was better in hemodialysis patients.

Among hemodialysis patients in the non-surgery group, 8 (21%) patients with lung cancer and 15 (56%) patients with liver cancer were considered to have resectable disease (Fig. 3 ). The reason patients with resectable cancer did not undergo surgery may be that less-invasive treatment options were available. Seven (88%) patients with lung cancer received radiation, and 10 (67%) patients with liver cancer received hepatic arterial infusion of chemotherapy (Fig. 2 ).

This study has several limitations. First, this is a retrospective study. Second, the possibility of selection bias existed. We enrolled cancer patients from designated cancer hospitals in Japan with organized oncology and nephrology departments, and did not enroll patients from other types of cancer hospitals. Therefore, our patients may have a better prognosis than hemodialysis cancer patients in other hospitals. Third, the follow-up period to evaluate the survival data was limited to 3 years, and relatively short. Fourth, unmeasured confounding factors associated with survival rate were not comprehensively collected, such as cancer stage, development of second cancer and performance status. Fifth, patients with less common cancers were not included in analysis. Sixth, the survival of hemodialysis patients is longer in Japan than in other countries, and it may impact the findings in the current study. However, the survival of the hemodialysis patients in Japan and in Western countries is reported to be similar in older patients [ 50 ]. Therefore, we believe that our study could be still informative in other countries. Finally, the data was limited to those diagnosed with cancer in between 2010 and 2012, and our data might not reflect the advancement of novel treatments, especially immunotherapy, adjuvant and neo-adjuvant therapies. A study with a more contemporary cohort would be warranted to understand the cancer outcomes of the patients with end-stage kidney disease.

In conclusion, despite these limitations, our study reports the clinical practice of cancer diagnosis and treatment for patients on hemodialysis. Our data suggest that the prognosis of hemodialysis cancer patients might be similar to that of the general cancer patients, and physicians should consider surgery a treatment option if dialysis patients are diagnosed with resectable cancer. Continued research is needed to develop appropriate cancer screening and treatment strategies for hemodialysis patients.

Supplementary Material

ACKNOWLEDGEMENTS

We especially thank the patients and their family members. We also thank all investigators and clinical research coordinators who contributed to this study at the 22 institutions, especially to Dr Naoya Kondo. Part of the work was presented in the abstracts for the ERA-EDTA Congress 2017 and ASN Kidney Week 2016.

Collaborators of the Onconephrology Consortium in Japan: Michio Nakamura, Kouichi Shiroshita, Mamiko Shimamoto (Sapporo City General Hospital); Toshikazu Moriwaki, Masahiro Hagiwara, Shuzo Kaneko, Kunihiro Yamagata (University of Tsukuba); Chihiro Kondo, Toshimi Takano, Noriko Hayami, Naoki Sawa, Yoshifumi Ubara (Toranomon Hospital); Naoki Izawa, Takuro Mizukami, Takashi Tsuda, Shina Sueki, Tsutomu Sakurada (St Marianna University School of Medicine); Naohiro Okano, Kazuhito Fukuoka, Shinya Kaname (Kyorin University); Keisei Taku, Haruomi Nishio, Noriko Mori (Shizuoka General Hospital); Tomoyuki Goto, Kosuke Ueda, Kazuyoshi Matsumura, Kunihiko Kosuga, Masataka Sasada (Shiga General Hospital); Ikuo Aoyama, Chiharu Kawanami, Yoko Shimizu, Sayako Maeda (Japanese Red Cross Otsu Hospital); Akira Nozaki, Kensei Yahata (National Hospital Organization Kyoto Medical Center); Akira Yoshioka, Shinji Ono (Mitsubishi Kyoto Hospital); Toshihiko Kirishima, Mayumi Tomita, Noriyuki Iehara (Kyoto City Hospital); Noriaki Tanaka, Chiharu Kinoshita (Kyoto Min-iren Chuo Hospital); Shigeki Kataoka, Yoshitaka Nishikawa, Shusuke Hiragi, Yuichiro Kitai, Shunsaku Nakagawa, Atsushi Yonezawa, Kazuo Matsubara (Kyoto University); Tadayuki Ko, Shujiro Yazumi, Tatsuo Tsukamoto, Eri Muso (Kitano Hospital); Takanobu Hayakumo, Toshikazu Takahashi (Takatsuki General Hospital); Masanori Toyoda, Kentaro Nakai, Shinichi Nishi (Kobe University); Takehiko Tsumura, Yukio Osaki, Keisuke Nishioka, Akira Sugawara (Red Cross Osaka Hospital); Hisateru Yasui, Akihiro Yoshimoto (Kobe City Medical Center General Hospital); Katsuyuki Tanabe, Tatsuyuki Inoue, Masashi Kitagawa (Okayama University); Eishi Baba, Toshiaki Nakano, Kazuhiko Tsuruya (Kyushu University); Soshi Terasaka, Koji Mitsuiki (Fukuoka Red Cross Hospital); Koichi Suyama, Takashige Kuwabara, Masashi Mukoyama (Kumamoto University); Masaomi Nangaku, Norio Hanafusa, Kenjiro Honda (University of Tokyo); Naoko Aragane, Hiroo Katsuya, Yohei Harada (Saga University).

FUNDING

This study was supported by Grants-in-Aid for Scientific Research (C) (16K09613 to T.M.), the Kidney Foundation of Japan (JKFB09-41 to T.M.) and Japanese Association of Dialysis Physicians (JADP Grant 2014-11 to T.M.).

AUTHORS’ CONTRIBUTIONS

Conceptualization: M.Y., N.T., T.M. Data curation: M.Y., N.T., S.Y., T.M., Y.K., K.S., S.Fukuma. Funding acquisition: M.Y., T.M. Project administration: M.Y. Formal analysis: N.T. Investigation: N.T., T.M. Supervision: M.Y. Writing—original draft: N.T., S.Y., T.M., Y.K., K.S. Writing—review and editing: M.Y., S.Y., T.F., T.H., T.Tsukamoto, N.M., K.D.J., S.Fukuma, T.Terada, M.M., S.Fukuhara.

DATA AVAILABILITY STATEMENT

Although the dataset that supports the findings of this study is not available because sharing of individual participant data with third parties was not specifically included in the informed consent of the study, data are available from the authors upon reasonable request and with permission of the ethical committee of Kyoto University Graduate School and Faculty of Medicine.

CONFLICT OF INTEREST STATEMENT

M.Y. receives research grants from Mitsubishi Tanabe Pharma and Boehringer Ingelheim. All other authors declare no competing interests.

REFERENCES

1. Maisonneuve P , Agodoa L , Gellert R et al. Cancer in patients on dialysis for end-stage renal disease: an international collaborative study . Lancet 1999 ; 354 : 93 – 9 . 10.1016/S0140-6736(99)06154-1 10408483

2. Iseki K , Osawa A , Fukiyama K . Evidence for increased cancer deaths in chronic dialysis patients . Am J Kidney Dis 1993 ; 22 : 308 – 13 . 10.1016/S0272-6386(12)70323-2 8352258

3. Vogelzang JL , van Stralen KJ , Noordzij M et al. Mortality from infections and malignancies in patients treated with renal replacement therapy: data from the ERA-EDTA registry . Nephrol Dial Transplant 2015 ; 30 : 1028 – 37 . 10.1093/ndt/gfv007 25637641

4. Masakane I , Taniguchi M , Nakai S et al. Annual Dialysis Data Report 2015, JSDT Renal Data Registry . Ren Replace Ther 2018 ; 4 : 19 . 10.1186/s41100-018-0149-8

5. Janus N , Launay-Vacher V , Thyss A et al. Management of anticancer treatment in patients under chronic dialysis: results of the multicentric CANDY (CANcer and DialYsis) study . Ann Oncol 2013 ; 24 : 501 – 7 . 10.1093/annonc/mds344 23038759

6. Reed M . Principles of cancer treatment by surgery . Surgery (Oxford) 2006 ; 24 : 70 – 3 . 10.1383/surg.2006.24.2.70

7. Miyamoto Y , Iwagami M , Aso S et al. Postoperative outcomes of cancer surgery in patients with and without kidney failure with dialysis therapy: a matched-pair cohort study . Clin Kidney J 2022 ; 15 : 1137 – 43 . 10.1093/ckj/sfac005 35664265 PMC9155240

8. Yeh CC , Lin JT , Jeng LB et al. Hepatic resection for hepatocellular carcinoma patients on hemodialysis for uremia: a nationwide cohort study . World J Surg 2013 ; 37 : 2402 – 9 . 10.1007/s00268-013-2137-z 23811792

9. Kida N , Morishima T , Tsubakihara Y et al. Stage at diagnosis and prognosis of colorectal, stomach, lung, liver, kidney, and bladder cancers in dialysis patients: a multicenter retrospective study using cancer registry data and administrative data . Nephron 2022 ; 146 : 429 – 38 . 35093954 10.1159/000521603

10. Otani Y , Okabayashi T , Shima Y et al. Safety and efficacy of the surgical management of hemodialysis patients with gastric cancer. Acta Med Okayama 2017 ; 71 : 333 – 9 . 10.18926/AMO/55310 28824189

11. Funakoshi T , Horimatsu T , Nakamura M et al. Chemotherapy in cancer patients undergoing haemodialysis: a nationwide study in Japan . ESMO Open 2018 ; 3 : e000301 . 10.1136/esmoopen-2017-000301 29531838 PMC5844381

12. Yamamoto H , Nishi S , Tomo T et al. 2015 Japanese Society for Dialysis Therapy: guidelines for renal anemia in chronic kidney disease . Ren Replace Ther 2017 ; 3 : 36 . 10.1186/s41100-017-0114-y

13. Wiedermann CJ , Wiedermann W , Joannidis M . Causal relationship between hypoalbuminemia and acute kidney injury . World J Nephrol 2017 ; 6 : 176 – 87 . 10.5527/wjn.v6.i4.176 28729966 PMC5500455

14. von Elm E , Altman DG , Egger M et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies . Int J Surg 2014 ; 12 : 1495 – 9 . 10.1016/j.ijsu.2014.07.013 25046131

15. Yoo KD , Lee JP , Lee SM et al. Cancer in Korean patients with end-stage renal disease: a 7-year follow-up . PLoS One 2017 ; 12 : e0178649 . 10.1371/journal.pone.0178649 28692695 PMC5503228

16. Chien CC , Han MM , Chiu YH et al. Epidemiology of cancer in end-stage renal disease dialysis patients: a national cohort study in Taiwan . J Cancer 2017 ; 8 : 9 – 18 . 10.7150/jca.16550 28123593 PMC5264035

17. Nakai S , Iseki K , Itami N et al. An overview of regular dialysis treatment in Japan (As of 31 December 2010) . Ther Apher Dial 2012 ; 16 : 483 – 521 . 10.1111/j.1744-9987.2012.01143.x 23190510

18. Nakai S , Watanabe Y , Masakane I et al. Overview of regular dialysis treatment in Japan (as of 31 December 2011) . Ther Apher Dial 2013 ; 17 : 567 – 611 . 10.1111/1744-9987.12147 24330555

19. Nakai S , Hanafusa N , Masakane I et al. An overview of regular dialysis treatment in Japan (as of 31 December 2012) . Ther Apher Dial 2014 ; 18 : 535 – 602 . 10.1111/1744-9987.12281 25523509

20. Locatelli F , Pisoni RL , Combe C et al. Anaemia in haemodialysis patients of five European countries: association with morbidity and mortality in the Dialysis Outcomes and Practice Patterns Study (DOPPS) . Nephrol Dial Transplant 2004 ; 19 : 121 – 32 . 10.1093/ndt/gfg458 14671047

21. Center for Cancer Control and Information Services NCC: the 2012 National Cancer Statistics Report from hospital-based Cancer registry . 2014 . https://ganjohojp/data/reg_stat/statistics/brochure/2012_reportpdf (16 August 2020, date last accessed).

22. Quint LE . Lung cancer: assessing resectability . Cancer Imaging 2003 ; 4 : 15 – 18 . 10.1102/1470-7330.2003.0028 18211855 PMC1434522

23. Masakane I , Nakai S , Ogata S et al. Annual Dialysis Data Report 2014, JSDT Renal Data Registry (JRDR) . Ren Replace Ther 2017 ; 3 : 18 .

24. Taborelli M , Toffolutti F , Del Zotto S et al. Increased cancer risk in patients undergoing dialysis: a population-based cohort study in North-Eastern Italy . BMC Nephrol 2019 ; 20 : 107 . 10.1186/s12882-019-1283-4 30922296 PMC6437907

25. Cheung CY , Chan GC , Chan SK et al. Cancer incidence and mortality in chronic dialysis population: a multicenter cohort study . Am J Nephrol 2016 ; 43 : 153 – 9 . 10.1159/000445362 27064839

26. Myung J , Choi JH , Yi JH et al. Cancer incidence according to the National Health Information Database in Korean patients with end-stage renal disease receiving hemodialysis . Korean J Intern Med 2020 ; 35 : 1210 – 9 . 10.3904/kjim.2018.400 32066223 PMC7487292

27. Hashimoto Y , Takagi T , Kondo T et al. Comparison of prognosis between patients with renal cell carcinoma on hemodialysis and those with renal cell carcinoma in the general population . Int J Clin Oncol 2015 ; 20 : 1035 – 41 . 10.1007/s10147-015-0812-9 25762166

28. Sugawara K , Yamashita H , Yajima S et al. Prognosis of hemodialysis patients undergoing surgery for gastric cancer: results of a multicenter retrospective study . Surgery 2021 ; 170 : 249 – 56 . 10.1016/j.surg.2021.01.014 33632543

29. Obuchi T , Hamanaka W , Yoshida Y et al. Clinical outcome after pulmonary resection for lung cancer patients on hemodialysis . Ann Thorac Surg 2009 ; 88 : 1745 – 8 . 10.1016/j.athoracsur.2009.08.010 19932229

30. Kang CH , Chen CH , Chiang PH . Primary urothelial carcinoma of the upper urinary tract in dialysis patients with 5-year follow-up . Jpn J Clin Oncol 2010 ; 40 : 241 – 6 . 10.1093/jjco/hyp143 19933687

31. Yossepowitch O , Sagy I , Margel D et al. Urothelial carcinoma of the bladder in patients on hemodialysis: clinical characteristics and oncological outcomes . J Urol 2012 ; 187 : 1215 – 9 . 10.1016/j.juro.2011.11.098 22335874

32. Hanafusa N , Nakai S , Iseki K et al. Japanese society for dialysis therapy renal data registry-a window through which we can view the details of Japanese dialysis population . Kidney Int Suppl 2015 ; 5 : 15 – 22 . 10.1038/kisup.2015.5 PMC4455188 26097781

33. Collins AJ , Foley RN , Gilbertson DT et al. United States Renal Data System public health surveillance of chronic kidney disease and end-stage renal disease . Kidney Int Suppl 2015 ; 5 : 2 – 7 . 10.1038/kisup.2015.2 PMC4455192 26097778

34. Wang JH , Hart A . Global perspective on kidney transplantation: United States . Kidney360 2021 ; 2 : 1836 – 9 . 10.34067/KID.0002472021 35373000 PMC8785833

35. Divard G , Goutaudier V . Global perspective on kidney transplantation: France . Kidney360 2021 ; 2 : 1637 – 40 . 10.34067/KID.0002402021 35372987 PMC8785777

36. Nitta K , Goto S , Masakane I et al. Annual dialysis data report for 2018, JSDT Renal Data Registry: survey methods, facility data, incidence, prevalence, and mortality . Ren Replace Ther 2020 ; 6 : 41 . 10.1186/s41100-020-00286-9

37. Han Y , Saran R . Global dialysis perspective: United States . Kidney360 2020 ; 1 : 1137 – 42 . 10.34067/KID.0001602020 35368785 PMC8815496

38. Hanafusa N , Fukagawa M . Global dialysis perspective: Japan . Kidney360 2020 ; 1 : 416 – 9 . 10.34067/KID.0000162020 35369370 PMC8809283

39. Goodkin DA , Bragg-Gresham JL , Koenig KG et al. Association of comorbid conditions and mortality in hemodialysis patients in Europe, Japan, and the United States: the Dialysis Outcomes and Practice Patterns Study (DOPPS) . J Am Soc Nephrol 2003 ; 14 : 3270 – 7 . 10.1097/01.ASN.0000100127.54107.57 14638926

40. Buccianti G , Maisonneuve P , Ravasi B et al. Cancer among patients on renal replacement therapy: a population-based survey in Lombardy, Italy . Int J Cancer 1996 ; 66 : 591 – 3 . 10.1002/(SICI)1097-0215(19960529)66:5<591::AID-IJC1>3.0.CO;2-O 8647617

41. Holley JL . Screening, diagnosis, and treatment of cancer in long-term dialysis patients . Clin J Am Soc Nephrol 2007 ; 2 : 604 – 10 . 10.2215/CJN.03931106 17699470

42. Shigeo N , Masaki O , Toru M et al. Cancer in patients undergoing chronic renal replacement therapy (in Japanese) . Jpn J Nephrol 2017 ; 59 : 610 – 4 .

43. Croswell JM , Ransohoff DF , Kramer BS . Principles of cancer screening: lessons from history and study design issues . Semin Oncol 2010 ; 37 : 202 – 15 . 10.1053/j.seminoncol.2010.05.006 20709205 PMC2921618

44. Woolf SH , Harris R . The harms of screening: new attention to an old concern . JAMA 2012 ; 307 : 565 – 6 . 10.1001/jama.2012.100 22318274

45. Kajbaf S , Nichol G , Zimmerman D . Cancer screening and life expectancy of Canadian patients with kidney failure . Nephrol Dial Transplant 2002 ; 17 : 1786 – 9 . 10.1093/ndt/17.10.1786 12270985

46. Rosner MH . Cancer screening in patients undergoing maintenance dialysis: who, what, and when . Am J Kidney Dis 2020 ; 76 : 558 – 66 . 10.1053/j.ajkd.2019.12.018 32305205

47. Mehrotra R , Davison SN , Farrington K et al. Managing the symptom burden associated with maintenance dialysis: conclusions from a Kidney Disease: Improving Global Outcomes (KDIGO) controversies conference . Kidney Int 2023 ; 104 : 441 – 54 . 10.1016/j.kint.2023.05.019 37290600

48. Caro JJ , Salas M , Ward A et al. Anemia as an independent prognostic factor for survival in patients with cancer: a systemic, quantitative review . Cancer 2001 ; 91 : 2214 – 21 . 10.1002/1097-0142(20010615)91:12<2214::AID-CNCR1251>3.0.CO;2-P 11413508

49. Esserman L , Shieh Y , Thompson I. Rethinking screening for breast cancer and prostate cancer . JAMA 2009 ; 302 : 1685 – 92 . 10.1001/jama.2009.1498 19843904

50. Nakai S , Wada A , Wakai K et al. Calculation of expected remaining lifetime of dialysis patients in Japan . Ren Replace Ther 2020 ; 6 : 58 . 10.1186/s41100-020-00301-z