This case series uses data from the National Cancer Database to evaluate trends in the characteristics, treatment, and outcomes of rectal adenocarcinoma in the US from 2004 to 2019.
Key Points
Question
What are the trends in the characteristics, treatment, and outcomes of rectal cancer in the US from 2004 to 2019?
Findings
In this case series study of 318 548 patients diagnosed with rectal adenocarcinoma, a significant increase was observed in the use of chemotherapy, immunotherapy, sphincter-saving surgery, and minimally invasive surgery along with increased time between diagnosis and surgery across 16 years from 2004 to 2019. These changes were associated with significant improvements in overall survival, reduced conversion rates, and shorter hospital stays.
Meaning
The findings of this study suggest that, despite increases in the numbers of patients diagnosed with advanced and metastatic rectal cancer, the overall survival has improved, probably due to improvements in therapeutic options and patient care.
Abstract
Importance
Rectal cancer management has significantly evolved over the last 2 decades.
Objective
This study aimed to evaluate trends in the characteristics, treatment, and outcomes of rectal adenocarcinoma across 16 years.
Design, Setting, and Participants
This retrospective, observational case series study used data from the National Cancer Database (NCDB) to evaluate patients diagnosed with rectal adenocarcinoma from 2004 through 2019. Data analysis was performed from March to May 2022.
Exposures
Trends in the treatment and outcomes of rectal adenocarcinoma in the US between 2004 and 2019 were explored. This period was subdivided into 4 equal periods: 2004-2007, 2008-2011, 2012-2015, and 2016-2019.
Main Outcomes and Measures
Patient and tumor characteristics, treatments, short-term outcomes, and overall survival.
Results
A total of 318 548 patients diagnosed with rectal adenocarcinoma were included in the analysis, 191 369 (60.1%) of whom were males and 127 179 (39.9%%) were females. The mean (SD) age of the patients was 63.5 (13.4) years, and 46 824 patients (14.8%) were younger than 50 years. Among the patients, 10 859 (3.4%) were of Asian race and ethnicity, 28 464 (8.9%) were Black, and 271 236 (85.1%) were White. The percentage of patients younger than 50 years who were diagnosed with rectal cancer increased by 1.5%, from 13.9% in period 1 to 15.4% in period 4. Patients in the last period (2016-2019) presented more often with stages III (36.2% vs 30.2% vs 25.0% vs 23.4%; P < .001) and IV (21.5% vs 19.3% vs 18.1% vs 18.6%; P < .001) disease compared with those in the remaining 3 periods. The use of chemotherapy (36.8% vs 48.1% vs 49.1% vs 47.0%; P < .001) and immunotherapy (0.4% vs 0.2% vs 3.5% vs 6.5%; P < .001) significantly increased across the 4 periods. Although neoadjuvant radiotherapy was used more often across the periods studied (28.6% in period 1 to 34.3% in period 4), the use of adjuvant radiotherapy was reduced by half (12.9% to 6.0%). The median (IQR) time from diagnosis to definitive surgery increased from 95 (15-126) days in period 1 to 128 (47-158) days in period 4. The rate of use of open surgery decreased by half (60.1% in period 2 to 30.1% in period 4), and the use of robotic surgery significantly increased (5.2% in period 2 to 28.4% in period 4). The conversion rate was significantly reduced (11.2% in period 2 to 7.3% in period 4) and the median (IQR) hospital stay decreased by 2 days, from 6 (3-9) days to 4 (2-7) days. The median (IQR) overall survival significantly increased across the periods (from 83.1 months [95% CI, 81.8-84.6 months] in period 1 to 92.1 months [95% CI, 90.2-93.6 months] in period 3; P < .001).
Conclusion and Relevance
The findings of this case series study suggest a treatment trend of increased use of chemotherapy, immunotherapy, sphincter-saving surgery, and minimally invasive surgery. In addition, the time between diagnosis and definitive surgery increased by a median of 33 days. This treatment trend was associated with a significant improvement in the overall survival, reduction in the conversion rate by 3.9%, and a 2-day shorter hospital stay. These findings have major clinical relevance to the management of rectal cancer. The improvements seen in short-term outcomes and survival of patients diagnosed with rectal cancer can probably be attributed to the treatment trends observed. Continued improvement in outcomes warrant further updates in treatments.
Introduction
Colorectal cancer (CRC) is the fourth most commonly diagnosed cancer in the world, with nearly 2 million new cases diagnosed every year.1 Rectal cancer accounts for 20% to 30% of the cases of CRC. However, the colon and rectum may differ in several aspects, including the embryologic origin, location, and surgical approach.2 Because the rectum is confined within the pelvis and in direct relation with the urinary bladder and genital organs, radiotherapy and surgery on the rectum are uniquely challenging.3
The management of rectal cancer has evolved substantially over the last years. Changes in treatment paradigms have been introduced and have been associated with improved overall and disease-free survival.4 Perhaps the most important breakthrough in the surgical treatment of rectal cancer was total mesorectal excision introduced in 1982.5 It has been proven that dissection in the embryological plane reduces local recurrence substantially.6 Another turning point was the increased use of minimally invasive surgery, which has improved the short-term outcomes without compromising oncologic outcomes or survival.7
The use of neoadjuvant chemoradiotherapy has also improved the outcomes of rectal cancer significantly.8 Two major clinical trials concluded the benefits of preoperative chemotherapy and radiotherapy in the treatment of rectal cancer.9,10 Over the last few years, the concept of total neoadjuvant therapy has also proved promising in achieving complete pathologic response and decreasing involved surgical margins.11 In addition, the use of immunotherapy in the treatment of advanced rectal cancer has showed promising preliminary results.12
We assessed the trends in the characteristics, treatment, and outcomes of rectal adenocarcinoma across more than 15 years by reviewing the National Cancer Database (NCDB).13 The NCDB includes hospital registry data from more than 1500 Commission on Cancer–accredited hospitals in the US and is a joint project between the Commission on Cancer of the American College of Surgeons and the American Cancer Society. The aim of our study was to assess whether a change in tumor characteristics and treatment methods was associated with a similar significant change in the outcomes, namely, survival.
Methods
Study Design, Setting, and Reporting
After obtaining access to the rectal cancer data from the NCDB (2004-2019), a retrospective analysis of patients diagnosed with rectal adenocarcinoma was conducted from March to May 2022. The study has been reported in line with the Strengthening the Reporting of Observational studies in Epidemiology (STROBE) guideline. Ethics/IRB approval was not necessary for our study as it was a retrospective review of a public data set that entailed deidentified patient data according to the common rule used for this type of study derived from public databases.
Study Population
The NCDB Participant User File was reviewed by two of us (S.E. and N.H.). The inclusion criteria were patients diagnosed with rectal adenocarcinoma (International Classification of Diseases for Oncology, Third Edition [ICD-O-3] code 8140/3, 8480-8481/3, 8490/3) recorded in the NCDB between 2004 and 2019. Both metastatic and nonmetastatic cases were eligible for inclusion, regardless of the treatment method. Patients with other tumor histology such as neuroendocrine tumors, sarcoma, melanoma, and gastrointestinal stromal tumors were excluded from the analysis.
Data Collection
The following patient data were collected and used for the analysis: age; sex; race and ethnicity (as classified in the NCDB to assess the trend of rectal cancer among different ethnicities); Charlson Comorbidity Index score; clinical and pathologic TNM (tumor, node, metastasis) stages; insurance status; geographic region; tumor histology and grade; lymphovascular invasion; number of lymph nodes examined; number of positive lymph nodes; receipt of chemotherapy, radiotherapy, and immunotherapy; sequencing of systemic therapy and radiotherapy; type and approach of surgery; and days from diagnosis to surgery and first treatment. The outcome data collected included conversion to open surgery, surgical margins, mortality, 30-day readmission, and overall survival.
A summary of the characteristics of the entire cohort of patients was described, after which the cohort was subdivided into 4 consecutive periods to assess the trends in the changes in patient demographics, tumor characteristics, treatments, and outcomes. Patients were subdivided into 4 equal periods to assess the trends in management and outcomes of rectal cancer.
Statistical Analysis
Statistical analyses were performed using EZR, version 1.55,14 R software, version 4.1.2 (The R Foundation), and SPSS, version 23 (IBM). Continuous data were expressed as mean and SD when normally distributed or otherwise as the median and IQR. An unpaired, 2-tailed t test or 1-way ANOVA test was used to analyze continuous variables. Categorical data were expressed in the form of numbers and percentages and were analyzed using the Fisher exact test or χ2 test. A complete case analysis approach was used to address missing data in the main outcomes. Kaplan-Meier statistics and log-rank tests were used to detect differences in 5-year overall survival between the periods. The receiver operating characteristic curve was used to identify the cutoff year for the changes in the main outcomes and treatment trends. The statistical threshold was 2-sided with P < .05 being considered statistically significant.
Results
Description of the Entire Cohort
After the exclusion of patients with other rectal cancer histology (eFigure 1 in the Supplement), a total of 318 548 patients diagnosed with rectal adenocarcinoma were included in the analysis. The mean (SD) age of the patients was 63.5 (13.4) years and 46 824 patients (14.8%) were younger than 50 years (Table 1). In addition, 191 369 patients (60.1%) were males and 127 179 (39.9%) were females. Among the patients, 1411 (0.4%) were American Indian, 10 859 (3.4%) were Asian, 28 464 (8.9%) were Black, 271 236 (85.1%) were White, and 3541 (1.1%) belonged to other races and ethnicities (Fiji Islander, Chamorro, Guamanian, Micronesian, Polynesian, Tahitian, Tongan, Samoan, and Melanesian). In all, 94.4% of patients resided in metropolitan or urban areas, 44.1% were insured by Medicare, and 41.9% had private insurance (Table 1). Of the tumors 95.1% were adenocarcinomas and 4.9% were mucinous or signet ring cell carcinomas. Among the patients, 6.2% had clinical TNM stage 0, 21.7% had stage I, 24.8% had stage II, 28.0% had stage III, and 19.2% had stage IV disease.
Table 1. Patient Characteristics and Pathologic Parameters Across 4 Periods (2004-2019).
| Characteristic | No. (%) | P value | |||
|---|---|---|---|---|---|
| 2004-2007 (n = 73 720) | 2008-2011 (n = 74 727) | 2012-2015 (n = 81 431) | 2016-2019 (n = 88 670) | ||
| Age, mean (SD), y | 64.7 (13.5) | 63.8 (13.6) | 63.0 (13.3) | 62.8 (13.1) | <.001 |
| Age category, y | |||||
| <50 | 10 142 (13.9) | 11 067 (14.9) | 12 075 (14.9) | 13 540 (15.4) | <.001 |
| ≥50 | 62 982 (86.1) | 63 012 (85.1) | 68 706 (85.1) | 74 487 (84.6) | |
| Males | 43 263 (58.7) | 44 389 (59.4) | 49 478 (60.8) | 54 239 (61.2) | <.001 |
| Females | 30 457 (41.3) | 30 338 (40.6) | 31 953 (39.2) | 34 431 (38.8) | |
| Race and ethnicity | |||||
| American Indian | 211 (0.3) | 325 (0.4) | 388 (0.5) | 487 (0.6) | <.001 |
| Asian | 1977 (2.7) | 2280 (3.1) | 2874 (3.6) | 3728 (4.2) | |
| Black | 5987 (8.2) | 6830 (9.2) | 7444 (9.2) | 8203 (9.3) | |
| White | 64 228 (88.1) | 63 950 (86.4) | 69 043 (85.5) | 74 015 (84.2) | |
| Othera | 470 (0.6) | 614 (0.8) | 1020 (1.2) | 1437 (1.7) | |
| Charlson Comorbidity Index score | |||||
| 0 | 57 390 (77.8) | 56 888 (76.1) | 61 925 (76.0) | 67 471 (76.1) | <.001 |
| 1 | 12 178 (16.5) | 13 111 (17.5) | 14 251 (17.5) | 13 299 (15.0) | |
| 2 | 3062 (4.2) | 3261 (4.4) | 3597 (4.4) | 4169 (4.7) | |
| 3 | 1090 (1.5) | 1467 (2.0) | 1658 (2.0) | 3731 (4.2) | |
| Geographic region | |||||
| Metropolitan | 57 702 (81.4) | 58 484 (81.0) | 63 734 (81.0) | 70 051 (81.2) | .003 |
| Urban | 11 455 (16.2) | 11 938 (16.5) | 13 191 (16.8) | 14 273 (16.6) | |
| Rural | 1688 (2.4) | 1761 (2.4) | 1803 (2.3) | 1903 (2.2) | |
| Insurance status | |||||
| Medicare | 34 747 (48.3) | 33 209 (45.4) | 34 764 (43.6) | 37 676 (43.0) | <.001 |
| Medicaid | 3084 (4.3) | 4492 (6.1) | 6224 (7.8) | 8069 (9.2) | |
| Other government | 865 (1.2) | 1084 (1.5) | 1228 (1.5) | 1495 (1.7) | |
| Private | 30 720 (42.7) | 31 201 (42.6) | 34 122 (42.8) | 37 441 (42.7) | |
| Not insured | 2460 (3.4) | 3208 (4.4) | 3444 (4.3) | 2932 (3.3) | |
| Facility location | |||||
| East North Central | 12 668 (18.4) | 12 995 (18.2) | 13 999 (17.8) | 16 240 (18.5) | <.001 |
| East South Central | 4788 (6.9) | 4790 (6.7) | 5406 (6.9) | 6411 (7.3) | |
| Middle Atlantic | 10 714 (15.5) | 11 148 (15.6) | 11 835 (15.0) | 13 160 (15.0) | |
| Mountain | 2936 (4.3) | 3227 (4.5) | 3486 (4.4) | 4064 (4.6) | |
| New England | 4104 (5.9) | 4217 (5.9) | 4542 (5.8) | 5125 (5.8) | |
| Pacific | 8410 (12.2) | 8692 (12.2) | 9660 (12.3) | 10 198 (11.6) | |
| South Atlantic | 13 834 (20.0) | 14 294 (20.0) | 16 268 (20.7) | 17 424 (19.8) | |
| West North Central | 5589 (8.1) | 5818 (8.2) | 6596 (8.4) | 7546 (8.6) | |
| West South Central | 5955 (8.6) | 6197 (8.7) | 6895 (8.8) | 7737 (8.8) | |
| Facility type | |||||
| Academic or research program | 21 733 (31.5) | 22 510 (31.5) | 24 829 (31.6) | 28 688 (32.6) | <.001 |
| Community cancer program | 5309 (7.7) | 5550 (7.8) | 6262 (8.0) | 6897 (7.8) | |
| Comprehensive community cancer program | 27 483 (39.8) | 28 732 (40.3) | 31 536 (40.1) | 34 395 (39.1) | |
| Integrated network cancer program | 14 473 (21.0) | 14 586 (20.4) | 16 060 (20.4) | 17 925 (20.4) | |
| Clinical TNM stage | |||||
| 0 | 3169 (8.2) | 3671 (6.5) | 3457 (5.3) | 2867 (4.2) | <.001 |
| 1 | 9118 (23.6) | 13 862 (24.6) | 13 449 (20.6) | 11 241 (16.3) | |
| 2 | 10 113 (26.2) | 14 523 (25.8) | 15 978 (24.5) | 15 107 (21.9) | |
| 3 | 9044 (23.4) | 14 093 (25.0) | 19 689 (30.2) | 24 947 (36.2) | |
| 4 | 7163 (18.6) | 10 200 (18.1) | 12 613 (19.3) | 14 828 (21.5) | |
| Histology | |||||
| Adenocarcinoma | 69 068 (93.7) | 70 804 (94.8) | 77 648 (95.3) | 85 517 (96.4) | <.001 |
| Mucinous carcinoma | 3973 (5.4) | 3328 (4.5) | 3143 (3.9) | 2549 (2.9) | |
| Signet ring cell carcinoma | 679 (0.9) | 595 (0.8) | 640 (0.8) | 604 (0.7) | |
| Grade | |||||
| Well differentiated | 6397 (10.7) | 6234 (10.3) | 6689 (10.3) | 3856 (11.0) | <.001 |
| Moderately differentiated | 43 723 (73.2) | 44 886 (74.2) | 49 414 (76.4) | 26 873 (76.5) | |
| Poorly differentiated | 9067 (15.2) | 8520 (14.1) | 7652 (11.8) | 3909 (11.1) | |
| Undifferentiated | 542 (0.9) | 845 (1.4) | 886 (1.4) | 488 (1.4) | |
| Lymphovascular invasion | |||||
| Yes | NA | 4374 (18.5) | 9610 (19.1) | 10 076 (20.3) | <.001 |
| No | NA | 19 285 (81.5) | 40 725 (80.9) | 39 466 (79.7) | |
| Pathologic TNM | |||||
| 0 | 3266 (7.2) | 2818 (6.3) | 3518 (7.0) | 2780 (7.1) | <.001 |
| 1 | 14 821 (32.9) | 14 916 (33.2) | 16 546 (32.8) | 12 401 (31.5) | |
| 2 | 9652 (21.4) | 9839 (21.9) | 10 824 (21.5) | 7627 (19.4) | |
| 3 | 12 473 (27.7) | 12 228 (27.2) | 13 933 (27.6) | 9954 (25.3) | |
| 4 | 4885 (10.8) | 5089 (11.3) | 5624 (11.1) | 6628 (16.8) | |
| Surgical margins | |||||
| Positive | 4168 (7.5) | 4452 (8.1) | 4553 (7.8) | 4538 (7.7) | <.001 |
| Negative | 51 413 (92.5) | 50 418 (91.9) | 53 532 (92.2) | 54 589 (92.3) | |
| No. of lymph nodes examined, median (IQR) | 11 (6-17) | 14 (8-19) | 15 (10-20) | 15 (12-21) | <.001 |
| No. of lymph nodes examined <12 | 23 577/39 718 (59.4) | 16 406/39 818 (41.2) | 14 279/42 196 (33.8) | 11 534/41 978 (27.5) | <.001 |
| No. of positive lymph nodes, median (IQR) | 0 (0-2) | 0 (0-1) | 0 (0-1) | 0 (0-1) | <.001 |
| Tumor size, median (IQR), mm | 37 (23-50) | 37 (22-50) | 40 (23-55) | 40 (24-56) | <.001 |
Abbreviations: NA, not available; TNM, tumor, node, metastasis.
Other races include: Fiji Islander, Chamorro, Micronesian, Guamanian, Polynesian, Tahitian, Tongan, Samoan, Melanesian.
During the study period, there have been 2 major updates in the American Joint Committee on Cancer’s TNM editions used in the NCDB. The sixth edition was used from 2004 to 2009, the seventh edition from 2010 to 2018, and the eighth edition in 2018 and 2019. There has been no significant change in data collection except during the eighth edition period, when the clinical and pathologic TNM stages were moved to dedicated columns, separate from the columns of TNM stages used in the earlier periods. To harmonize this discrepancy, we copied the clinical and pathologic TNM stages from their dedicated 2018-2019 columns to the original columns where TNM stages of all previous years were stored, thus having a complete data set of TNM staging for all years. The median overall survival for the entire cohort was 90.15 months (95% CI, 89.40-90.94 months).
Patient and Tumor Characteristics Across Periods
Patients were divided into 4 equal periods: period 1 (2004-2007), period 2 (2008-2011), period 3 (2012-2015), and period 4 (2016-2019). Period 1 included 73 720 patients; period 2, 74 727 patients; period 3, 81 431 patients, and period 4, 88 670 patients. Compared with the first period, there was a 20.3% increase in the number of patients in the last period.
As given in Table 1, the mean (SD) age at diagnosis of rectal adenocarcinoma decreased by 2 years, from 64.7 (13.5) years in the first period to 62.8 (13.1) years in the last period. The percentage of patients with early-onset rectal cancer (before age 50 years) increased by 1.5%, from 13.9% in period 1 to 15.4% in period 4. From the first to the last period, there was a slight increase in male patients (58.7% to 61.2%), Black patients (8.2% to 9.3%), and Asian patients (2.7% to 4.2%). More patients with severe comorbidities (Charlson Comorbidity Index score of 3) were included in period 4 (4.2% vs ≤2.0% in the previous 3 periods). The number of patients with insurance provided by Medicaid doubled, from 4.3% in period 1 to 9.2% in period 4. Patients in period 4 presented with a more locally advanced (stage III) and metastatic (stage IV) disease than patients in the previous periods.
A decrease in the percentage of patients diagnosed with mucinous carcinoma (5.4% to 2.9%) and poorly differentiated adenocarcinomas (15.2% to 11.1%) was noted from period 1 to period 4. The rates of positive resection margin increased from 7.5% to 7.7%. The median (IQR) number of harvested lymph nodes increased from 11 (6-17) in the first period to 15 (10-20) in the third period and 15 (12-21) in the last period (Table 1).
Changes in Treatment Trends
When the first and last periods were compared, the use of chemotherapy and immunotherapy significantly increased (chemotherapy: odds ratio [OR], 1.52; 95% CI, 1.49-1.55; P < .001; immunotherapy: OR, 19.55; 95% CI, 17.25-22.20; P < .001) (Figure 1). The use of combined neoadjuvant and adjuvant chemotherapies for nonmetastatic rectal cancer increased from 10.8% in period 1 to 19.7% in period 4. Although neoadjuvant radiotherapy was used more often over time (28.6% in period 1 to 34.3% in period 4) (OR, 1.29; 95% CI, 1.26-1.31; P < .001), the use of adjuvant radiotherapy was reduced by half (12.9% to 6.0%) (OR, 0.43; 95% CI, 0.41-0.44; P < .001) (Table 2 and Figure 2).
Figure 1. Trends in the Use of Immunotherapy (2004-2019).
Table 2. Treatment and Outcome Trends Across 4 Periods (2004-2019).
| Factor | No. (%) | P value | |||
|---|---|---|---|---|---|
| 2004-2007 (n = 73 720) | 2008-2011 (n = 74 727) | 2012-2015 (n = 81 431) | 2016-2019 (n = 88 670) | ||
| Chemotherapy | |||||
| Yes | 26 630 (36.8) | 35 554 (48.1) | 39 772 (49.1) | 41 427 (47.0) | <.001 |
| No | 45 689 (63.2) | 38 401 (51.9) | 41 154 (50.9) | 46 704 (53.0) | |
| Immunotherapy | |||||
| Yes | 256 (0.4) | 148 (0.2) | 2841 (3.5) | 5760 (6.5) | <.001 |
| No | 71 846 (99.6) | 73 856 (99.8) | 78 452 (96.5) | 82 669 (93.5) | |
| Sequencing of systemic treatment in patients without metastases | |||||
| No systemic therapy | 7380 (44.7) | 20 587 (45.1) | 22 186 (42.5) | 11 653 (42.9) | <.001 |
| Neoadjuvant | 5659 (34.3) | 14 003 (30.7) | 16 221 (31.1) | 8133 (29.9) | |
| Adjuvant | 1683 (10.2) | 4853 (10.6) | 4516 (8.6) | 2040 (7.5) | |
| Neoadjuvant and adjuvant | 1774 (10.8) | 6210 (13.6) | 9292 (17.8) | 5341 (19.7) | |
| Intraoperative | 3 (0.004) | 4 (0.005) | 9 (0.01) | 6 (0.007) | |
| Intraoperative with preoperative or postoperative systemic treatment | 0 | 9 (0.01) | 15 (0.02) | 7 (0.008) | |
| Sequencing of radiation | |||||
| No radiotherapy given | 40 410 (54.8) | 39 881 (53.4) | 43 814 (53.8) | 50 522 (57.0) | <.001 |
| Neoadjuvant | 21 093 (28.6) | 25 078 (33.6) | 29 097 (35.7) | 30 417 (34.3) | |
| Adjuvant | 9476 (12.9) | 7126 (9.5) | 5839 (7.2) | 5336 (6.0) | |
| Neoadjuvant and adjuvant | 196 (0.3) | 215 (0.3) | 245 (0.3) | 234 (0.3) | |
| Intraoperative | 7 (0.009) | 16 (0.02) | 19 (0.02) | 10 (0.01) | |
| Intraoperative with preoperative or postoperative radiotherapy | 27 (0.04) | 29 (0.04) | 44 (0.1) | 62 (0.1) | |
| Time from diagnosis to first treatment, median (IQR), d | 21 (3-36) | 25 (7-40) | 27 (11-43) | 32 (15-49) | <.001 |
| Time from diagnosis to radiotherapy, median (IQR), d | 38 (24-63) | 37 (20-60) | 39 (27-61) | 48 (32-90) | <.001 |
| Time from diagnosis to surgery in patients without metastases who underwent surgery, median (IQR), d | 95 (15-126) | 102 (23-133) | 118 (36-146) | 128 (47-158) | <.001 |
| Surgery conducted | |||||
| Yes | 50 875 (69.1) | 50 102 (67.1) | 52 649 (64.7) | 52 503 (59.2) | <.001 |
| No | 22 777 (30.9) | 24 559 (32.9) | 28 694 (35.3) | 36 126 (40.8) | |
| Type of surgery | |||||
| Low anterior resection | 25 323 (49.8) | 25 657 (51.2) | 27 448 (52.1) | 28 647 (54.6) | <.001 |
| Abdominoperineal resection | 8411 (16.5) | 8177 (16.3) | 8577 (16.3) | 7825 (14.9) | |
| Local excision | 11 157 (21.9) | 10 294 (20.5) | 10 453 (19.9) | 10 525 (20) | |
| Pull through coloanal anastomosis | 2743 (5.4) | 2712 (5.4) | 2903 (5.5) | 2645 (5.1) | |
| Pelvic exenteration | 976 (1.9) | 939 (1.9) | 1110 (2.1) | 1256 (2.4) | |
| Total proctocolectomy | 1019 (2) | 1111 (2.2) | 1165 (2.2) | 960 (1.8) | |
| Proctectomy NOS | 319 (0.6) | 342 (0.7) | 300 (0.6) | 194 (0.4) | |
| Surgery NOS | 927 (1.8) | 870 (1.7) | 693 (1.3) | 451 (0.9) | |
| Approach | |||||
| Open or not specified | NA | 15 070 (60.1) | 23 325 (44.3) | 15 808 (30.1) | <.001 |
| Laparoscopic | NA | 8698 (34.7) | 22 116 (42.0) | 21 752 (41.4) | |
| Robotic | NA | 1306 (5.2) | 7201 (13.7) | 14 930 (28.4) | |
| Conversiona | |||||
| Yes | NA | 1122 (11.2) | 2767 (9.4) | 2695 (7.3) | <.001 |
| No | NA | 8882 (88.8) | 26 550 (90.6) | 33 987 (92.7) | |
| Total | NA | 10 004 | 29 317 | 36 682 | |
| 30-d Mortalityb | |||||
| Yes | 894 (1.8) | 744 (1.5) | 727 (1.4) | 429 (1.1) | <.001 |
| No | 49 533 (98.2) | 49 124 (98.5) | 51 674 (98.6) | 39 381 (98.9) | |
| Total | 50 427 | 49 868 | 52 401 | 39 810 | |
| 90-d Mortalityb | |||||
| Yes | 1843 (3.7) | 1549 (3.1) | 1371 (2.6) | 858 (2.2) | <.001 |
| No | 48 446 (96.3) | 48 221 (96.9) | 50 901 (97.4) | 38 597 (97.8) | |
| Total | 50 289 | 49 770 | 52 272 | 39 455 | |
| 30-d Readmissionb | |||||
| No readmission | 44 642 (91.1) | 45 015 (91.3) | 48 496 (92.6) | 48 733 (93.3) | <.001 |
| Unplanned readmission | 2937 (6.0) | 3192 (6.5) | 3109 (5.9) | 2894 (5.5) | |
| Planned readmission | 1423 (2.9) | 1074 (2.2) | 757 (1.4) | 603 (1.2) | |
| Total | 49 002 | 49 281 | 52 362 | 52 230 | |
| Hospital stay, median (IQR), db | 6 (3-9) | 6 (3-8) | 5 (3-7) | 4 (2-7) | <.001 |
| Duration of radiotherapy, median (IQR), d | 29 (0-40) | 32 (0-40) | 33 (0-40) | 0 (0-39) | <.001 |
| Follow-up, median (IQR), mo | 68.4 (22.3-143.7) | 68.6 (23.5-115.6) | 56.5 (23.3-74.9) | 30.3 (17.9-41.9) | <.001 |
Abbreviations: NA, not available; NOS, not otherwise specified.
Outcome assessed only in patients who underwent laparoscopic or robotic resections with known conversion status.
Outcome assessed only in patients who underwent surgery with known outcome status.
Figure 2. Trends in the Use of Neoadjuvant and Adjuvant Chemotherapy and Radiotherapy (2006-2019).
The median (IQR) time from diagnosis to first treatment increased by 11 days, from 21 (3-36) days in period 1 to 32 (15-49) days in period 4, while the median (IQR) time from diagnosis to definitive surgery increased from 95 (15-126) days in the first period to 128 (47-158) days in the last period; this was associated with a trend of increased survival. Abdominoperineal resection was used less often (OR, 0.82; 95% CI, 0.79-0.85; P < .001) whereas the percentage of patients who did not undergo surgery increased (OR, 1.54; 95% CI, 1.51-1.57; P < .001). The approach to surgery significantly changed as the rate of open surgery decreased significantly (60.1% in period 2 to 30.1% in period 4) (OR, 0.28; 95% CI, 0.28-0.29; P < .001), and the rate of robotic surgery increased significantly (5.2% in period 2 to 28.4% in period 4) (OR, 7.23; 95% CI, 6.82-17.67; P < .001) (Figure 3).
Figure 3. Trends in the Surgical Approach Used (2010-2019).
Change in Outcome Trends
The conversion rate significantly decreased from 11.2% in period 2 to 7.3% in period 4 (OR, 0.63; 95% CI, 0.58-0.67; P < .001) for patients who underwent laparoscopic or robotic resections. There were no missing data in conversion to open surgery. There was a slight reduction in the rates of 30-day mortality (1.8% to 1.1%; P < .001) and 90-day mortality (3.7% to 2.2%; P < .001) and 30-day unplanned readmission (6.0% to 5.5%; P < .001) for patients who underwent surgery. The proportion of missing data was 7.0% for 30-day and 90-day mortality and 2.0% for readmission. The median (IQR) hospital stay decreased by 2 days from 6 (3-9) days in period 1 to 4 (2-7) days in period 4 (Table 2).
Using Kaplan-Meier statistics, the median 5-year overall survival significantly increased across the first 3 periods, whereas the last period was not included in the analysis as it did not have a sufficient length of follow-up (83.1 months [95% CI, 81.8-84.6 months] in period 1, 87.8 months [95% CI, 86.6-90.3 months] in period 2, and 92.1 months [95% CI, 90.2-93.6 months] in period 3, P < .001) (eFigure 2 in the Supplement). The percentage of patients with fewer than 12 harvested nodes significantly decreased, from 59.4% in the first period to 27.5% in the last period.
Cutoff Years for Change in Trends
The cutoff year for days from diagnosis to surgery was 2011 (sensitivity = 58.1%, specificity = 46.9%); for the use of chemotherapy, 2009 (sensitivity = 51.8%, specificity = 62.1%); for the use of immunotherapy, 2013 (sensitivity = 94.3%, specificity = 53.7%); and for robotic surgery, 2015 (sensitivity = 74.5%, specificity = 54.6%). The cutoff year for a change in trends for overall survival was 2012 (sensitivity = 68.7%, specificity = 54.2%); for 30-day mortality, 2009 (sensitivity = 45.8%, specificity = 60.9%); for 90-day mortality, 2011 (sensitivity = 60.3%, specificity = 48.2%); and for hospital stay, 2007 (sensitivity = 82.9%, specificity = 19.1%).
Discussion
The present study explored trends for changes in the characteristics, treatments, and outcomes of rectal adenocarcinoma in the US over a period of 16 years. There has been a 20.3% increase in the incidence of rectal cancer over this period. It has been reported that the incidence of rectal cancer has been rising faster than that of colon cancer, with an annual increase of 2% from the 1990s to 2013.15 The rate of early-onset rectal cancer showed an increase of 1.5% between 2004 and 2019. It has been previously reported that the incidence of early-onset CRC in the US has been increasing since the mid-1990s. Data from the Surveillance, Epidemiology, and End Results database showed that the incidence of early-onset CRC in the US doubled between 1991 and 2014.15 An interesting observation was that the percentage of patients diagnosed with rectal cancer who had severe comorbidities almost tripled, perhaps attesting to the improved health care services and resources, enabling better management of this group of patients.
The pathologic parameters of rectal cancer also showed some interesting changes. A decrease was seen in certain types of adenocarcinoma that are typically associated with a poorer prognosis, such as mucinous carcinomas and poorly differentiated adenocarcinomas. Mucinous carcinomas usually present in a more locally advanced stage, may not show a good response to neoadjuvant therapy, and may have poor prognosis in younger patients, females, and patients with T4 tumors.16,17 Another observation was the increase in patients with stage IV disease. This observation emphasizes the pivotal role of CRC screening that aims to protect against CRC by detecting precancerous lesions before they evolve into cancer and by detecting early asymptomatic cancers.18
The number of harvested lymph nodes significantly increased over time. This finding is clinically relevant since the number of examined lymph nodes is associated with the overall survival of patients with CRC.19 A minimum of 12 harvested lymph nodes was considered one of the key quality metrics for the treatment of CRC.20 The percentage of patients with fewer than 12 harvested nodes significantly decreased, from 59.4% in the first period to 27.5% in the last period. This finding may reflect an improved surgical practice with the implementation of standardized techniques.
Significant changes in the treatment paradigms of rectal cancer were seen over the years. The use of chemotherapy considerably increased 1.5-fold across the years. The cutoff year for a remarkable increase in the use of chemotherapy was 2009, the same year when a clinical trial found that the addition of chemotherapy after the initial neoadjuvant chemoradiotherapy was associated with considerably higher rates of complete response with acceptable toxicity rates.21 In addition, a Cochrane review22 published in 2009 concluded that combined preoperative chemotherapy and radiotherapy enhanced the pathological response and improved local control in stages II and III rectal cancer compared with preoperative radiotherapy alone.
Similarly, an increase of almost 20-fold in the use of immunotherapy was observed. Immunotherapy of CRC has been regarded as a promising option for CRC treatment.23 A phase 2 trial concluded that pembrolizumab provided an objective response rate of 40% and progression-free survival rate of 78% in CRC patients with mismatch repair-deficient status.24
The surgical treatment of rectal cancer has also undergone remarkable change as the use of abdominoperineal resection decreased in favor of sphincter-saving surgery. This observation might be related to the increased performance of rectal cancer surgery in specialized, high-volume centers with sufficient resources and adequate surgeon expertise that allow more sphincter-saving procedures to be performed. Indeed, a study that examined the association of volume with outcome in rectal cancer surgery found that the shift toward higher volume was associated with better overall outcomes.25
Another important observation was the reduction in the use of open surgery by 50% (from 60.1% in period 2 to 30.1% in period 4), which is conversely commensurate with the increased adoption of minimally invasive surgery. This fact was particularly obvious with robotic surgery, for which the uptake was increased by more than 5-fold. The rise in robotic surgery use may be explained by the claimed benefits of this platform.26 The year 2015 marked the cutoff for the increased uptake of robotic surgery use, the same year when the ALaCaRT trial was published and could not establish the noninferiority of laparoscopic compared with open surgery for patients with T1 to T3 rectal cancers.27
Although the increase in time between diagnosis and first treatment was statistically significant over time, it may not be clinically significant. In all cases, the median time before treatment was initiated was less than 60 days, which is consistent with the recommendation by the National Accreditation Program for Rectal Cancer.28,29
The aforementioned changes in treatment paradigms and the increased time before surgery were associated with a significant increase in overall survival. This notable improvement in survival, despite the increased prevalence of patients with stage IV disease, was attestable to better multidisciplinary care and improved therapeutic options. The introduction of the National Accreditation Program for Rectal Cancer may offer further insights into outcome trends in the future.29
Strengths and Limitations
Although the treatment trends noted in the present study would be anticipated, there has been little documentation of the competing treatment paradigms in the literature, which is a point of strength for this study. However, a number of limitations, including the retrospective nature of the study and lack of data on disease recurrence, should be noted. The study’s very large sample size may result in a type I error as some statistically significant results may not be clinically meaningful; thus, the differences in trends should be carefully interpreted. An important limitation is that our study was based on the NCDB data set, which was developed as a joint quality improvement program of the Commission on Cancer between the American College of Surgeons and the American Cancer Society.30 The NCDB encompasses approximately 70% of all newly diagnosed cancers in the US31; however, since the NCDB is hospital based rather than population based, it does not represent the US population overall, and thus generalizability from these data is more limited than that of other databases.32 Future studies using data from nonaccredited US hospitals are needed to compare the trends reported in our study with those from different databases. Last, survival reported in this study is the overall survival, which is not specific to cancer. However, this may explain the lower survival rates in the first period, as some patients may have died because of aging or other, non–cancer-related causes.
Conclusions
The findings of this case series study showed that a treatment trend of increased use of chemotherapy, immunotherapy, sphincter-saving surgery, and minimally invasive surgery was noted between 2004 and 2016. In addition, the time between diagnosis and definitive surgery increased. This treatment trend was associated with a significant improvement in overall survival, a reduction in conversion rate by 3.9%, and a 2-day shorter hospital stay.
Although there were statistically significant increases in early-onset rectal cancer, rectal cancer affecting patients of Black and Asian races and ethnicities, and rectal cancer affecting male patients, these differences were marginal and thus may not be clinically meaningful. The findings of the study are of major clinical relevance to the management of rectal cancer. The improvements seen in short-term outcomes and survival of patients diagnosed with rectal cancer are probably attributable to the treatment trends observed. Continued improvement in outcomes warrant further updates in treatments.
eFigure 1. Flow Chart for Patient Selection for the Study
eFigure 2. Kaplan-Meier Survival Curve Across the 4 Time Periods
References
- 1.Rawla P, Sunkara T, Barsouk A. Epidemiology of colorectal cancer: incidence, mortality, survival, and risk factors. Prz Gastroenterol. 2019;14(2):89-103. doi: 10.5114/pg.2018.81072 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Wei EK, Giovannucci E, Wu K, et al. Comparison of risk factors for colon and rectal cancer. Int J Cancer. 2004;108(3):433-442. doi: 10.1002/ijc.11540 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Fazeli MS, Keramati MR. Rectal cancer: a review. Med J Islam Repub Iran. 2015;29:171. [PMC free article] [PubMed] [Google Scholar]
- 4.Keller DS, Berho M, Perez RO, Wexner SD, Chand M. The multidisciplinary management of rectal cancer. Nat Rev Gastroenterol Hepatol. 2020;17(7):414-429. doi: 10.1038/s41575-020-0275-y [DOI] [PubMed] [Google Scholar]
- 5.Heald RJ, Husband EM, Ryall RDH. The mesorectum in rectal cancer surgery—the clue to pelvic recurrence? Br J Surg. 1982;69(10):613-616. doi: 10.1002/bjs.1800691019 [DOI] [PubMed] [Google Scholar]
- 6.Arbman G, Nilsson E, Hallböök O, Sjödahl R. Local recurrence following total mesorectal excision for rectal cancer. Br J Surg. 1996;83(3):375-379. doi: 10.1002/bjs.1800830326 [DOI] [PubMed] [Google Scholar]
- 7.Cheong C, Kim NK. Minimally invasive surgery for rectal cancer: current status and future perspectives. Indian J Surg Oncol. 2017;8(4):591-599. doi: 10.1007/s13193-017-0624-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Zhao F, Wang J, Yu H, et al. Neoadjuvant radiotherapy improves overall survival for T3/4N+M0 rectal cancer patients: a population-based study of 20300 patients. Radiat Oncol. 2020;15(1):49. doi: 10.1186/s13014-020-01497-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Sauer R, Becker H, Hohenberger W, et al. ; German Rectal Cancer Study Group . Preoperative versus postoperative chemoradiotherapy for rectal cancer. N Engl J Med. 2004;351(17):1731-1740. doi: 10.1056/NEJMoa040694 [DOI] [PubMed] [Google Scholar]
- 10.Cedermark B, Dahlberg M, Glimelius B, Påhlman L, Rutqvist LE, Wilking N; Swedish Rectal Cancer Trial . Improved survival with preoperative radiotherapy in resectable rectal cancer. N Engl J Med. 1997;336(14):980-987. doi: 10.1056/NEJM199704033361402 [DOI] [PubMed] [Google Scholar]
- 11.Kasi A, Abbasi S, Handa S, et al. Total neoadjuvant therapy vs standard therapy in locally advanced rectal cancer: a systematic review and meta-analysis. JAMA Netw Open. 2020;3(12):e2030097. doi: 10.1001/jamanetworkopen.2020.30097 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Golshani G, Zhang Y. Advances in immunotherapy for colorectal cancer: a review. Therap Adv Gastroenterol. Published online June 1, 2020. [DOI] [PMC free article] [PubMed]
- 13.Hopewood P. Using the NCDB to explore trends in cancer. Bulletin of the American College of Surgeons. 2017. Accessed April 26, 2022. https://bulletin.facs.org/2017/07/using-the-ncdb-to-explore-trends-in-cancer/ [PubMed]
- 14.Kanda Y. Investigation of the freely available easy-to-use software “EZR” for medical statistics. Bone Marrow Transplant. 2013;48(3):452-458. doi: 10.1038/bmt.2012.244 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Siegel RL, Fedewa SA, Anderson WF, et al. Colorectal cancer incidence patterns in the United States, 1974-2013. J Natl Cancer Inst. 2017;109(8):109. doi: 10.1093/jnci/djw322 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Madbouly KM, Mashhour AN, Omar W. Is it safe to omit neoadjuvant chemo-radiation in mucinous rectal carcinoma? Int J Surg. 2015;23(pt A):120-127. [DOI] [PubMed]
- 17.Emile SH, Magdy A, Elnahas W, Hamdy O, Abdelnaby M, Khafagy W. Predictors for local recurrence and distant metastasis of mucinous colorectal adenocarcinoma. Surgery. 2018;164(1):56-65. doi: 10.1016/j.surg.2017.11.028 [DOI] [PubMed] [Google Scholar]
- 18.Emile SH, Barsom SH, Wexner SD. An updated review of the methods, guidelines of, and controversies on screening for colorectal cancer. Am J Surg. 2022;224(1 Pt B):339-347. doi: 10.1016/j.amjsurg.2022.03.034 [DOI] [PubMed] [Google Scholar]
- 19.Compton CC, Fielding LP, Burgart LJ, et al. Prognostic factors in colorectal cancer: College of American Pathologists Consensus Statement 1999. Arch Pathol Lab Med. 2000;124(7):979-994. doi: 10.5858/2000-124-0979-PFICC [DOI] [PubMed] [Google Scholar]
- 20.Swanson RS, Compton CC, Stewart AK, Bland KI. The prognosis of T3N0 colon cancer is dependent on the number of lymph nodes examined. Ann Surg Oncol. 2003;10(1):65-71. doi: 10.1245/ASO.2003.03.058 [DOI] [PubMed] [Google Scholar]
- 21.Habr-Gama A, Perez RO, Sabbaga J, Nadalin W, São Julião GP, Gama-Rodrigues J. Increasing the rates of complete response to neoadjuvant chemoradiotherapy for distal rectal cancer: results of a prospective study using additional chemotherapy during the resting period. Dis Colon Rectum. 2009;52(12):1927-1934. doi: 10.1007/DCR.0b013e3181ba14ed [DOI] [PubMed] [Google Scholar]
- 22.Ceelen WP, Van Nieuwenhove Y, Fierens K. Preoperative chemoradiation versus radiation alone for stage II and III resectable rectal cancer. Cochrane Database Syst Rev. 2009;(1):CD006041. doi: 10.1002/14651858.CD006041.pub2 [DOI] [PubMed] [Google Scholar]
- 23.Koido S, Ohkusa T, Homma S, et al. Immunotherapy for colorectal cancer. World J Gastroenterol. 2013;19(46):8531-8542. doi: 10.3748/wjg.v19.i46.8531 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Le DT, Uram JN, Wang H, et al. PD-1 blockade in tumors with mismatch-repair deficiency. N Engl J Med. 2015;372(26):2509-2520. doi: 10.1056/NEJMoa1500596 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Siragusa L, Sensi B, Vinci D, et al. Volume-outcome relationship in rectal cancer surgery. Discov Oncol. 2021;12(1):11. doi: 10.1007/s12672-021-00406-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Chouhan H, Shin J, Kim SH. Is robotic rectal resection the preferred option for resectable cancer? Mini-invasive Surgery. 2018;2:18. doi: 10.20517/2574-1225.2018.40 [DOI] [Google Scholar]
- 27.Stevenson AR, Solomon MJ, Lumley JW, et al. ; ALaCaRT Investigators . Effect of laparoscopic-assisted resection vs open resection on pathological outcomes in rectal cancer: the ALaCaRT Randomized Clinical Trial. JAMA. 2015;314(13):1356-1363. doi: 10.1001/jama.2015.12009 [DOI] [PubMed] [Google Scholar]
- 28.Wexner SD, Berho ME. The rationale for and reality of the New National Accreditation Program for Rectal Cancer. Dis Colon Rectum. 2017;60(6):595-602. doi: 10.1097/DCR.0000000000000840 [DOI] [PubMed] [Google Scholar]
- 29.National Accreditation Program for Rectal Cancer Standards Manual. Accessed May 5, 2022. https://www.facs.org/quality-programs/cancer-programs/national-accreditation-program-for-rectal-cancer
- 30.Boffa DJ, Rosen JE, Mallin K, et al. Using the National Cancer Database for outcomes research: a review. JAMA Oncol. 2017;3(12):1722-1728. doi: 10.1001/jamaoncol.2016.6905 [DOI] [PubMed] [Google Scholar]
- 31.American College of Surgeons . About the National Cancer Database. 2019. Accessed April 9, 2019. https://www.facs.org/quality-programs/cancer/ncdb/about
- 32.Jairam V, Park HS. Strengths and limitations of large databases in lung cancer radiation oncology research. Transl Lung Cancer Res. 2019;8(suppl 2):S172-S183. doi: 10.21037/tlcr.2019.05.06 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eFigure 1. Flow Chart for Patient Selection for the Study
eFigure 2. Kaplan-Meier Survival Curve Across the 4 Time Periods