Abstract
Lymph nodes with acellular mucin harvested from treated colorectal cancers (CRC) are staged as pN0. However, there is variability among pathologists while reporting pN stage when acellular mucin is found within nodes of untreated CRCs. While the UICC guidelines suggest staging them as pN1, the AJCC and CAP do not offer any recommendations. In order to characterize their clinicopathologic features and outcome, we compared 16 untreated CRCs (study group; mean age: 68 yrs) harboring nodes with acellular mucin with 34 pN0 and 25 pN1 untreated CRC controls. All tumors were unifocal; 12 (75%) were right-sided lesions. Most cases (75%) showed one node with acellular mucin (range: 1–3). MMR-deficient tumors were significantly more common in the study group (83%) compared to pN0 (33%; p=0.006) and pN1 controls (8%; p<0.001). The overall survival of study group patients was closer to pN0 compared to pN1 controls; however, this difference was not statistically significant. In conclusion, untreated CRC that harbor acellular mucin within lymph nodes commonly present as right-sided, MMR-deficient tumors in older women that show a non-mucinous phenotype. While the limited number of cases precludes us from making any formal recommendations about staging, we suggest that the finding of acellular mucin in a node should prompt evaluation of deeper levels (with or without cytokeratin immunohistochemistry) and submission of all pericolonic fat for additional lymph node harvest. Whether acellular mucin in nodes of untreated CRCs is related to the indolent biology of the disease, a robust local immune response or MMR-deficiency requires further investigation.
Keywords: Acellular mucin, lymph nodes, colon cancer, untreated
INTRODUCTION
Lymph node status provides important prognostic information for patients undergoing resection for colorectal cancer (CRC). Patients with node-positive or high-risk stage II disease who receive adjuvant therapy have a distinct survival advantage over those who do not. [1–3] In contrast to colon cancer, neoadjuvant chemoradiation followed by surgical resection is the standard of care for treating locally advanced rectal cancer. [4] The American Joint Committee on Cancer (AJCC) [5] and College of American Pathologists (CAP) guidelines [6] recommend that lymph nodes with acellular mucin found in neoadjuvantly treated rectal cancer resections should be staged as pN0. [7] However, no such guidelines exist for reporting nodal stage when acellular mucin is found within nodes from treatment-naïve CRC resections. With regards to European guidelines, the frequently asked questions section of the Tumor Node Metastasis (TNM) supplement of Union for International Cancer Control (UICC) staging for colon cancer states that - “ If there has been no neoadjuvant therapy, mucin pools are considered positive for tumour”. [8] As a result, there is variability in reporting practices, such that some pathologists categorize them as negative for metastasis, while others stage them as positive nodes.
In an effort to expand our understanding regarding untreated CRCs that harbor acellular mucin within lymph nodes, we decided to study the clinicopathologic characteristics of this group of tumors and compare their clinical outcome with two control groups consisting of treatment naïve node-positive and node-negative CRCs.
MATERIALS AND METHODS
Study population
Pathology databases from three institutions (Cleveland Clinic, Massachusetts General Hospital and Memorial Sloan Kettering Cancer Center) were queried for the terms “acellular, “mucinous” and “mucin pools” in all CRC resections reviewed between 1990 to 2020. For a combined inhouse resection volume of 20, 617 across three institutions, we identified 11 intramural cases of untreated CRCs and 5 extramural consults with lymph nodes containing acellular mucin, without any additional positive nodes. Thus 16 cases formed the final study group. By definition, patients who received any prior neoadjuvant therapy were excluded from the study. Four cases had lymph nodes with acellular mucin (range: 1–3) along with additional positive nodes (range: 1–10). Since these additional positive nodes would influence the survival data and/or clinical decision about adjuvant therapy, these cases were excluded from our analysis. Controls were chosen from a consecutive series of inhouse resections specimens and consisted of 59 treatment naïve CRCs matched for age, tumor location, and pathologic tumor stage (pT). They were further divided into node-negative (pN0; 34) and node-positive groups (pN1=25).
Clinicopathological Data
Patient age and gender, tumor location, tumor size, gross appearance, number of tumors, tumor grade/differentiation, percentage of mucinous component, small vessel lymphovascular invasion (LVI), perineural space invasion (PNI), extramural venous invasion (EMVI), tumor deposits (TDs), primary tumor stage (pT), number of lymph nodes with acellular mucin, total number of lymph nodes harvested, mismatch repair protein (MMR) expression status (immunohistochemical [IHC] stain results), BRAF/KRAS mutation status (if available) were recorded from patient records and following histologic review. Staging parameters (pN and pM) and overall prognostic Stage Group were re-scored according to current (8th ed.) AJCC Cancer Staging Manual guidelines. [5] Tumors located from the cecum to the distal transverse colon inclusive were considered to be right-sided tumors, whereas those arising in the splenic flexure and distally were categorized as left-sided tumors. Tumors exhibiting a mucin content of ≥ 50% were classified as mucinous adenocarcinomas. As a standard sampling procedure, all tumors were sampled adequately by obtaining at least one full-thickness tumor section for every cm of tumor. Additional deeper levels performed in all 16 cases, and cytokeratin immunohistochemistry performed in 8/16 cases. Patient outcomes were evaluated by reviewing medical records (including physician notes, imaging studies, etc.) and information regarding administration of adjuvant therapy (if any), disease progression (defined as loco-regional recurrence and/or metachronous development of distant metastases), and disease specific mortality was noted. Follow-up time was defined as months from the date of index colectomy to death or last clinical follow-up. The study protocol was approved by the respective Institutional Review Boards (Protocol numbers: Cleveland Clinic −15–1636; Massachusetts General Hospital - 2017P000061; Memorial Sloan-Kettering Cancer Center - GI 16–1682).
Immunohistochemistry
Immunohistochemistry for MMR proteins was performed on formalin-fixed paraffin-embedded 4 micron thick sections using mouse anti-MLH1 monoclonal antibody (1:100 dilution; clone ES05; Novocastra, Buffalo Grove, IL), mouse anti-PMS2 monoclonal antibody (1:50 dilution; clone MRQ-28; Cell Marque, Rocklin, CA), mouse anti-MSH2 monoclonal antibody (1:150 dilution, clone FE11; Calbiochem, San Diego, CA), and mouse anti-MSH6 monoclonal antibody (1:50 dilution, clone PU29;Novocastra) using the Envision Plus Detection System (Dako, Carpinteria, CA). Cases were characterized as MMR-deficient if tumor cells showed loss of staining for any of the four proteins, with intact staining in adjacent tissue (i.e., an internal positive control was required).
Statistical analysis
Study variables and overall survival for cases (study group) were compared separately to pN0 and pN1 controls. Categorical variables were described as frequency counts and percentages and compared with Chi-square test or Fisher’s exact test; missing responses were noted but not used to calculate percentages. Continuous variables were described as mean and standard deviation or median and range and compared with Wilcoxon rank sum test. Overall survival was estimated with Kaplan-Meier method and compared with log-rank test. Data were analyzed with SAS® software (SAS Institute, Inc., Cary, NC, USA). All statistical tests were two-sided; p <0.05 was used to indicate statistical significance.
RESULTS
Clinical and Pathologic Characteristics of Study Group
The clinical and pathologic features of the study group are summarized in Table 1. The mean age of the study group patients (10 women, 6 men) was 68 years (median 66; range: 51–87). All were unifocal tumors; 12 (75%) were located in the right colon (11 ascending colon/cecum and 1 transverse colon), while 4 were left-sided tumors. The median tumor size was 5.7 cm (range: 2.0 – 9.4). The size of invasive component could not be assessed accurately in Case 14 since this specimen was received as a piecemeal polypectomy for a malignant cecal polyp. This patient subsequently underwent right hemicolectomy, which showed no evidence of residual tumor, but exhibited acellular mucin in one of twenty-two lymph nodes. Nearly two-thirds of the tumors were in the form of ulcerated lesions (8/13; 62%), while 5/13 (38%) presented as polypoid masses. The pT stage distribution was as follows: pT1 – 1 (6%), pT2 – 4 (25%), pT3 – 10 (63%) and pT4 – 1 (6%). A little over half the lesions were moderately differentiated adenocarcinoma (56%); 25% were poorly differentiated and 19% were well-differentiated adenocarcinomas. Only 3/16 (20%) cases were categorized as invasive mucinous adenocarcinoma based on their mucinous content. Lymphovascular invasion (LVI) or foci suspicious for LVI were present in 8 (50%) cases; all showed small vessel invasion. There was no evidence of extramural venous invasion in any of the cases. Interestingly, 10/12 (83%) cases in which MMR protein immunohistochemistry was performed showed an MMR-deficient status; all 10 showed loss of MLH-1 expression. Among the 4 cases in whom MMR IHC could not be performed, 3 showed Crohn’s like lymphoid response, tumor infiltrating lymphocytes, and/or peritumoral lymphoid infiltrate suggestive of an MMR-deficient profile. Lynch syndrome was excluded, either on the basis of presence of BRAF V600E mutation /MLH-1 promoter hypermethylation or by performing a detailed review of clinical and family history. Thus, of the 10 cases with MLH-1 loss, BRAF V600E mutation was found in all 4 cases that underwent BRAF mutation testing and MLH-1 hypermethylation was noted in one additional case. Review of medical records did not reveal any relevant family or personal history that would suggest Lynch syndrome in the remaining 5 cases. KRAS mutation analysis was only performed in one case (case 15), which showed a wild-type status.
TABLE 1:
Clinical and Pathologic characteristics of cases with acellular mucin within lymph nodes (Study group; n=16)
| Case No. | Age/Gender | Anatomic location | Size (cm) | Grade | % mucinous component | LVI | pT stage | Total no. of lymph nodes retrieved | No. of nodes with acellular mucin | Ancillary stains or deeper sections performed | Original interpretation of LN with acellular mucin | MMR status | Follow-up duration, status |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 71/M | Descending colon | 2.5 | G2 | <50% | No | pT2 | 18 | 1 | Deeper levels and AE1/AE3 IHC | Negative | Deficient, MLH-1 loss | 15 mo, NED |
| 2 | 56/M | Ascending colon | 5 | G2 | <50% | No | pT3 | 89 | 1 | Deeper levels | Positive | Unknown | Unknown |
| 3 | 79/F | Descending colon | 9 | G2 | ≥50% | No | pT4 | 27 | 2 | Deeper levels | Negative | Deficient, MLH-1 loss | 15 mo, Died of unrelated cause |
| 4 | 66/F | Ascending colon | 6 | G2 | ≥50% | No | pT3 | 16 | 1 | Deeper levels, AE1/AE3 IHC | Negative | Deficient, MLH-1 loss | 65 mo, NED |
| 5 | 76/F | Ascending colon | 3 | G2 | <50% | Yes | pT2 | 21 | 2 | AE1/AE3 IHC | Negative | Deficient, MLH-1 loss | 48 mo, Died of unrelated cause |
| 6 | 80/F | Ascending colon | 5 | G2 | <50% | No | pT3 | 14 | 1 | Deeper levels and AE1/AE3 IHC | Negative | Unknown | Unknown |
| 7 | 66/M | Ascending colon | 3.8 | G3 | <50% | Yes | pT3 | 35 | 3 | Deeper levels | Positive | Deficient MLH-1 loss | 170 mo, NED |
| 8 | 84/F | Ascending colon | 8 | G3 | <50% | Suspicious | pT3 | 17 | 1 | Deeper levels and AE1/AE3 IHC | Negative | Deficient MLH-1 loss | 221 mo, NED |
| 9 | 71/F | Ascending colon | 2 | G3 | <50% | Yes | pT2 | 18 | 1 | Deeper levels | Positive | Unknown | 189 mo, Died of unrelated cause |
| 10 | 59/M | Ascending colon | 6.3 | G2 | <50% | No | pT3 | At least 3* | 1 | Deeper levels | Positive | Unknown | Unknown |
| 11 | 58/F | Ascending colon | 4.5 | G3 | <50% | No | pT2 | 29 | 1 | Deeper levels, AE1/AE3, IHC | Comment: Possibility of a positive node cannot be excluded | Deficient; MLH-1 loss | 146 mo, NED |
| 12 | 87/F | Ascending colon | 6 | G2 | <50% | Yes | pT3 | 18 | 2 | Deeper levels | Positive | Deficient, MLH-1 loss | 47 mo, NED |
| 13 | 61/F | Descending colon | 5.7 | G1 | <50% | Yes | pT3 | 20 | 1 | Deeper levels, AE1/AE3 IHC | Positive | Deficient, MLH-1 loss | 2mo, NED |
| 14 | 62/M | Ascending colon | Unknown** | G1 | Unknown | No | pT1** | 22 | 1 | Deeper levels, AE1/AE3 IHC | Positive | Deficient, MLH-1 loss | 60 mo, NED |
| 15 | 53/F | Rectosigmoid colon | 5.8 | G2 | ≥50% | Yes | pT3 | 25 | 1 | Deeper levels | Positive | Proficient# | <1 mo, NED |
| 16 | 51/M | Transverse colon | 9.4 | G1 | <50% | Yes | pT3 | 42 | 1 | Deeper levels | Positive | Proficient | Unknown |
M: Male, F: Female, NED: No evidence of disease, PNI: Perineural invasion; LVI: Lymphovascular invasion; MMR: Mismatch repair
International consult case; per report, multiple lymph nodes were examined; however, total lymph node count was not available while reviewing the case
Malignant colon polyp resected by polypectomy followed by right hemicolectomy; original size of invasive carcinoma could not be recorded; select slides reviewed as a consult
KRAS mutation performed; wild-type status; Cases 3, 4, 5, 12 - BRAFV600E mutation detected
The median number of lymph nodes harvested was 20 (range 3 – 89). The total number of lymph nodes with acellular mucin ranged from 1–3 (Figure 1), with a majority of the cases (12; 75%) containing only one node with acellular mucin. Additional deeper levels, with or without cytokeratin immunohistochemistry (IHC), were performed in all 16 cases and did not reveal any epithelial cells.
Figure 1:
Acellular mucin deposit within a pericolonic lymph node from a 66-year-old woman with invasive moderately differentiated adenocarcinoma involving the ascending colon (pT3). (20X magnification)
Follow-up information was available in 12/16 (75%) cases and their median follow-up period was 45 months (range: 1 – 220). Of these, 9 (75%) are alive with no evidence of disease and 3 (25%) died of an unrelated cause. As expected, there was variability in assigning the final lymph node status within our study cohort where 6/16 cases were called negative, 9/16 were assigned a node-positive status and the remaining case was signed out with a comment indicating that the possibility of a positive node could not be excluded. Among the 6 cases which were called node-negative by the original pathologist, 2 (cases 1 and 4) did not receive adjuvant therapy and are alive without evidence of disease (follow-up duration: 15 and 65 mo, respectively); 3 (cases 3, 5 and 8) died of an unrelated cause and 1 (case 6) was lost to follow-up. Among the 9 cases that were called positive, one received chemotherapy but was lost to follow-up after the 2 month visit and one was recommended adjuvant therapy at a local hospital but did not follow-up after the 1 month visit. Two additional patients did not receive therapy and are alive without disease (follow-up duration 46 and 96 mo); the remaining 5 patients were either lost to follow-up (n = 3) or lacked information regarding adjuvant therapy (n = 2). Case 11 was that of a pT2 ascending colon cancer where the pathology report included a comment indicating that the possibility of a positive lymph node could not be excluded. This patient is alive and well with no evidence of disease after a follow-up duration of 146 months.
Comparison of Clinical and Pathologic Features Between Study Group and Control Groups
Table 2 shows the comparative analysis between study group and control group patients. A majority of the patients in the study group were women (63%) while there was an almost equal male to female distribution in the two control groups patients (Table 2). A significantly higher percentage of tumor in the study group were poorly differentiated (25%) compared to those in pN0 (9%; p=0.025) and pN1 groups (0%; p=0.007). Both study group and pN0 control group cases showed small vessel invasion whereas extramural venous invasion was only seen in pN1 group. The study group tumors were significantly more likely to be MMR-deficient (83%) compared to both pN0 (33%; p=0.006) and pN1 controls (8%; p<0.001). There were no differences in the rest of the parameters evaluated between the study group and control groups.
TABLE 2:
Comparison of clinical and pathologic features between the study group and controls groups
| Variable | Study group (N=16) | pN0 controls (N=34) | pN1 controls (N=25) | P value Study group vs pN0 controls |
P value Study group vs pN1 controls |
|---|---|---|---|---|---|
|
| |||||
| Age (y) | |||||
| Mean ± SD | 68 ± 11 | 65 ± 13 | 66 ± 14 | NA* | NA* |
| Median (range) | 66 (51–87) | 66 (40–90) | 66 (35–87) | ||
|
| |||||
| Gender | |||||
| Male (%) | 6 (37.5%) | 16 (47.1%) | 12 (48.0%) | 0.53 | 0.51 |
| Female (%) | 10 (62.5%) | 18 (52.9%) | 13 (52.0%) | ||
|
| |||||
| Tumor location | |||||
| Right | 12 (75.0%) | 25 (73.5%) | 20 (80.0%) | NA* | NA* |
| Left | 4 (25.0%) | 9 (26.5%) | 5 (20.0%) | ||
|
| |||||
| Gross Appearance | |||||
| Polypoid | 4 (30.8%) | 16 (48.5%) | 14 (56.0%) | 0.21 | 0.16 |
| Ulcerated | 8 (61.5%) | 17 (51.5%) | 11 (44.0%) | ||
| Both | 1 (7.7%) | 0 (0%) | 0 (0%) | ||
| Unknown | 3 | 1 | 0 | ||
|
| |||||
| Tumor size (cm) | (N=15) | ||||
| Mean ± SD | 5.5 ± 2.2 | 4.4 ± 1.6 | 4.3 ± 2.4 | 0.07 | 0.10 |
| Median (range) | 5.7 (2.0–9.4) | 4.5 (1.6–8.5) | 3.3 (1.3–10.0) | ||
|
| |||||
| Histologic grade | |||||
| G1 Well | 3 (18.8%) | 19 (55.9%) | 14 (56.0%) | 0.025 | 0.007 |
| G2 Moderate | 9 (56.2%) | 12 (35.3%) | 11 (44.0%) | ||
| G3 Poor | 4 (25.0%) | 3 (8.8%) | 0 (0%) | ||
|
| |||||
| Mucinous differentiation | |||||
| <50% | 12 (80.0%) | 24 (70.6%) | 22 (88.0%) | 0.73 | 0.65 |
| ≥50% | 3 (20.0%) | 10 (29.4%) | 3 (12.0%) | ||
| Unknown | 1 | 0 | 0 | ||
|
| |||||
| Lymphovascular invasion | |||||
| No | 8 (50.0%) | 25 (73.5%) | 9 (36.0%) | 0.10** | 0.37** |
| Suspicious | 1 (6.2%) | 0 (0%) | 0 (0%) | ||
| Yes | 7 (43.8%) | 9 (26.5%) | 16 (64.0%) | ||
| Small vessel Only | 8 (100%) | 9 (100%) | 10 (62.5%) | - | 0.07 |
| Small and EMVI | 0 (0%) | 0 (0%) | 6 (37.5%) | ||
|
| |||||
| Perineural invasion | |||||
| No | 15 (100.0%) | 13 (100.0%) | 6 (85.7%) | - | 0.32 |
| Yes | 0 (0%) | 0 (0%) | 1 (14.3%) | ||
| Unknown | 1 | 21 | 18 | ||
|
| |||||
| Median total no. of LN (range) | 20 (3–89) | 24 (10–89) | 29 (7–57) | 0.54 | 0.12 |
|
| |||||
| pT stage | |||||
| T1 | 1 (6.2%) | 0 (0%) | 0 (0%) | NA* | NA* |
| T2 | 4 (25.0%) | 9 (26.5%) | 8 (32.0%) | ||
| T3 | 10 (62.5%) | 24 (70.6%) | 16 (64.0%) | ||
| T4 | 1 (6.2%) | 1 (2.9%) | 1 (4.0%) | ||
|
| |||||
| Mismatch repair status | |||||
| Proficient | 2 (16.7%) | 22 (66.7%) | 23 (92.0%) | 0.006 | <0.001 |
| Deficient | 10 (83.3%) | 11 (33.3%) | 2 (8.0%) | ||
| Unknown | 4 | 1 | 0 | ||
|
| |||||
| Adjuvant therapy | |||||
| No | 8 (72.7%) | 26 (81.2%) | 12 (54.5%) | 0.67 | 0.46 |
| Yes | 3 (27.3%) | 6 (18.8%) | 10 (45.5%) | ||
| Unknown | 5 | 2 | 3 | ||
|
| |||||
| Median follow-up, mo (range) | 54 (1–220) | 56 (1–175) | 35 (1–182) | 0.95 | 0.55 |
|
| |||||
| Follow-up status | |||||
| Alive - no evidence of disease | 9 (75.0%) | 21 (61.8%) | 14 (56.0%) | 0.20 (log-rank test)# | 0.14 (log-rank test)# |
| Died of disease | 0 (0%) | 1 (2.9%) | 1 (4%) | ||
| Died of unrelated causes | 3 (25.0%) | 10 (29.4%) | 8 (32.0%) | ||
| Died of unknown causes | 0 (0%) | 2 (5.9%) | 2 (8.0%) | ||
| Unknown | 4 | 0 | 0 | ||
LN - lymph node; NA - not applicable; EMVI- extramural venous invasion
Study group and controls were matched for age, anatomic location and pT stage
Categories of “Yes” and “suspicious” for lymphovascular were combined for this analysis
Cases (n=2) and controls (pN0; n=1) who are alive with no evidence of disease, but with a short follow-up (less than 6 months) duration were excluded from the survival analysis
Outcome analysis
Patients who were found to be alive without evidence of disease but had very short follow-up (<6 months), were excluded from the survival analysis. Thus, 2 cases from the study group (<1 month follow-up each) and 1 pN0 control group patient (3 months) were excluded. Figure 2 shows that the overall survival of the study group patients was more similar to pN0 compared to pN1 controls; however, this comparison did not achieve statistical significance. Upon restricting the analysis to MMR-D tumors in all three groups, we found that the survival of patients harboring acellular mucin was closer to pN0 patients compared to pN1 patients (Supplementary figure 1).
Figure 2:
Kaplan-Meier curve comparing survival outcomes shows that the clinical outcome of patients with acellular mucin (cases; study group) was more similar to pN0 controls compared to pN+ (N1) colon cancer controls.
DISCUSSION
Lymph node status is a known predictor of overall and disease-free survival in patients with colorectal cancer such that those with no nodal disease (stage I and II) have a 5-year survival rate of greater than 75% compared to the 30% - 60% survival rate in node-positive stage III and IV disease. [9, 10] Positive nodal disease is therefore considered an indication for adjuvant chemotherapy. [11]
Prior studies have shown that acellular pools of mucin in specimens following neoadjuvant therapy do not have any impact on local recurrence or patient survival, and therefore should be considered to represent completely eradicated tumor. [7, 12–14] A study by Smith et al. showed that in 11/27 patients where acellular mucin was found at the margins or within nodes, there was no impact on local recurrence or outcome. [7] Based on these studies, treated nodes with acellular mucin are staged as pN0 disease. Unlike treated rectal cancer, UICC and AJCC/CAP differ in their recommendations for reporting nodal stage when acellular mucin is found in CRCs without prior therapy. The UICC system suggests staging them as pN+ disease, while AJCC/CAP offer no such recommendation.
A recently published web-based questionnaire collated reporting practices related to challenging areas in colorectal cancer staging from a diverse group of 118 surgical pathologists. One of the questions posed to the pathologists was whether lymph nodes containing acellular mucin should be classified as node-negative or node-positive disease.[15] Nearly two-thirds of the respondents staged acellular mucin in lymph nodes from untreated CRCs as pN0. These results are quite similar to results from another survey in which greater than 90% of North American pathologists considered such cases to represent pN0 disease. [16] European pathologists were more likely to stage acellular mucin deposits in lymph nodes as pN1 based on UICC guidelines. Our results are in agreement with these surveys. In our series of 16 cases, 6 (38%) were categorized as pN0, 9 (56%) were called pN1 and one was reported as “cannot exclude positive node”. It is likely that the group of pathologists who categorize these nodes as negative believe that epithelial cells need to be present in tissue sections in order to call these nodes positive, while those who categorize them as pN+ nodes argue that mucin within a lymph node has to originate from the primary tumor/glandular epithelium, even when it cannot be demonstrated on the slide. The findings from the surveys as well as our study underscore the need for uniform reporting guidelines to ensure consistent staging and post-operative management.
Our study is the first to demonstrate that acellular mucin within lymph nodes is most commonly associated with right-sided colonic adenocarcinomas that arise in women in their 6th decade of life (median age: 68 y). Most cases tend to be well or moderately differentiated tumors that present as pT3 lesions and harbor an MMR-deficient profile. The total number of lymph nodes that contain acellular mucin ranges between one to three nodes, with most cases containing just one node with this finding. Of the 16 cases, majority of the patients are alive without evidence of disease (n=9) or died of unrelated causes (n=4). Only two patients either received or were recommended adjuvant chemotherapy. Upon comparing their survival with age, location and pT stage-matched controls, we found that the overall survival of study group patients with acellular mucin was closer to pN0 controls than pN1 controls, although this result was not statistically significant. The lack of significant differences in outcome is likely due to the limited number of cases in our study cohort.
Interestingly, unlike the prevailing assumption that mucin within lymph nodes is most likely to originate from tumors that have a significant (≥50%) mucinous component, we found that only three out of the sixteen cases qualified as invasive mucinous adenocarcinomas based on their mucin content. These results are quite similar to the three cases reported in the literature (see Table 3). [17, 18] Two of these three cases were documented in older women (85 y and 72 y) with right-sided colon cancers. Two were pT3 tumors while the third was a pT4 tumor. Unlike our series, two of the three reported tumors were categorized as invasive mucinous adenocarcinomas.
TABLE 3:
Clinicopathologic features of three cases of colorectal cancers with acellular mucin within lymph nodes reported in the literature
| Cases | Age/Gender | Anatomic location | TNM Stage | %Mucin content | Total number of lymph nodes with acellular mucin | Clinical outcome |
|---|---|---|---|---|---|---|
|
| ||||||
| Foong KS et al13 | ||||||
| Case 1 | 85/F | Hepatic flexure | T4 N0 M0 | ≥50% | 2/21 | No adjuvant therapy offered due to age |
| Case 2 | 57/M | Rectum | T3 N0 M0 | 30% | 1/28 | Unknown |
|
| ||||||
| Gandhi J et al14 | 72/F | Cecum | T3 N0 M0 | ≥50% | 2/15 | Adjuvant therapy given; no follow-up available |
An important observation noted in our study is that tumors with acellular mucin were more likely to be MMR-deficient (83%). Studies have shown that sporadic MMR-deficient CRCs are associated with older age group, female gender, right-sided location, [19] and carry a better overall prognosis, stage for stage, compared with sporadic MMR-proficient CRCs. [20] Therefore, it is quite possible that predominance of MMR-deficient tumors in the study group may independently contribute to the clinical outcome in this cohort. In order to explore this further, we performed a sub-analysis comparing outcome among MMR-D tumors in all three cohorts and found that the survival profile of patients harboring acellular mucin was closer to pN0 patients than pN1 patients (Supplementary figure 1). This data suggests that MMR-D status may not be the only factor influencing better outcome in the study group. We also attempted a multivariable analysis to assess the influence of this variable and other relevant clinical and pathologic features (grade, LVI, PNI) on patient outcome; however, the limited number of patients in the study group precluded a meaningful analysis. Thus, although patient survival in our cohort was found to be closer to node-negative disease, thereby supporting a pN0 status, the limited number of cases precludes formal recommendation regarding staging these tumors.
We are not entirely sure why right-sided, MMR-D CRCs in older women with a non-mucinous phenotype would be prone to harboring acellular mucin within lymph nodes. While it seems logical that mucin within a lymph node must originate from the primary colon cancer via lymphatic spread, we were unable to demonstrate accompanying epithelial cells in any of our cases. This is similar to previously reported cases. It is difficult to envision how mucin could reach the lymph node parenchyma without being transported by metastatic epithelium. One potential explanation could be a robust immune response that has managed to destroy the presumably low-volume of tumor cells. This perhaps explains the favorable outcome of such cases which was found to be more similar to ‘pN0’ disease. Whether this finding is related to the indolent biology of the disease, a robust immune response or MMR-deficient status requires further investigation. To that end, additional studies, including studying the immune milieu and corresponding molecular characteristics of these lesions using newer sequencing technologies will be helpful towards this end.
Although our study represents the largest cohort of untreated colon cancers with acellular mucin in nodes, it has several limitations that are noteworthy. Firstly, despite collating cases from three large volume practices, the overall number of the study cases is low. Given that the cases also included extramural consults, we were unable to obtain clinical follow-up for some of these cases. Having said that, the median follow-up duration for the study group was more than three years (54 mo), an interval that is widely accepted as a surrogate for assessing disease-free survival in patients with CRC. In cases that had adequate follow-up, these patients were all monitored in accordance with the standard of care for post-operative management of CRC prevalent at that point of time. Secondly, the strict matching process for the series of 59 controls chosen for our analysis raises the possibility of underrepresenting all CRCs resected at our institutions. However, all controls were chosen from a consecutive series of inhouse resection specimens. These matching criteria were included to eliminate the influence of potential confounders such as age, location and pT stage on patient outcome. Therefore, we believe that a selection bias is less likely to be introduced during this process. Thirdly, as all cases were collated at tertiary care centers, with one-third being consultation cases, referral bias is a possibility. However, since this is such a rare finding, it would be difficult to collect a large series and perform a meaningful analysis without including these consult cases. While these cases would certainly merit review by an expert in colorectal cancer, we are hopeful that our study not only provides guidance to surgical pathologists about how to approach these cases but also provides some objective data about outcome. Lastly, although multiple deeper sections with cytokeratin IHC were performed in all the nodes with this finding, it is difficult to recommend a universal practice guideline about number of deeper sections to evaluated and/or use of cytokeratin IHC. In the study published by Kirsch et al, 90% of surveyed pathologists performed additional tissue levels and/or cytokeratin IHC. It is possible that we may have found rare epithelial cells by completely exhausting the blocks. However, this is not a very practical approach and is unlikely to be employed by most surgical pathologists. We certainly recommend that the finding of acellular mucin in a node should prompt evaluation of deeper tissue levels and submission of all pericolonic fat for additional lymph node evaluation.
In summary, the findings from our series of treatment-naïve colorectal cancers with acellular mucin in lymph nodes highlights some important clinicopathologic associations. This phenomenon is commonly associated with right-sided tumors which occur in older women. The tumors tend to be well or moderately differentiated adenocarcinomas that display a non-mucinous phenotype and are frequently MMR-deficient. The number of lymph nodes exhibiting this finding usually ranges between one and three. Patients with adequate follow-up information appear to have good clinical outcome without locoregional recurrence or distant metastasis. A prospective study that includes a larger cohort of patients is needed to validate our findings and to assess the role of adjuvant therapy (including immunotherapy), if any, in this setting.
Supplementary Material
Supplementary Figure 1: Kaplan-Meier curve comparing survival outcomes of all MMR-D tumors in three groups shows that the clinical outcome of patients with acellular mucin (cases; study group) was more similar to pN0 controls compared to pN+ (N1) colon cancer controls. The number of cases and events in this sub-analysis were too low to calculate p values.
Footnotes
DECLARATIONS:
Conflict of interest: Alok A. Khorana, M.D. - Receives Institutional grant support from NIH, CDC, Leap, BMS, Celgene, Merck; Consulting fees from Janssen, Bayer, Pfizer, Anthos, Sanofi, BMS, Seagen, honorarium from Medscape for CME events, Support for attending meetings and/or travel from Janssen, Bayer, Seagen, BMS; Participated on a Data
Safety Monitoring Board or Advisory Board - Bayer, BMS, Janssen; Medical and Scientific Advisory Board, NBCA (unpaid).
All remaining authors have no conflicts of interest.
Data Availability Statement:
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
REFERENCES:
- 1.Booth CM, Nanji S, Wei X, Peng Y, Biagi JJ, Hanna TP, Krzyzanowska MK, Mackillop WJ (2016) Use and Effectiveness of Adjuvant Chemotherapy for Stage III Colon Cancer: A Population-Based Study J Natl Compr Canc Netw 14:47–56. doi: 10.6004/jnccn.2016.0006 [DOI] [PubMed] [Google Scholar]
- 2.Haller DG, Catalano PJ, Macdonald JS, O’Rourke MA, Frontiera MS, Jackson DV, Mayer RJ (2005) Phase III study of fluorouracil, leucovorin, and levamisole in high-risk stage II and III colon cancer: final report of Intergroup 0089 J Clin Oncol 23:8671–8678. doi: 10.1200/JCO.2004.00.5686 [DOI] [PubMed] [Google Scholar]
- 3.Sargent D, Sobrero A, Grothey A, O’Connell MJ, Buyse M, Andre T, Zheng Y, Green E, Labianca R, O’Callaghan C, Seitz JF, Francini G, Haller D, Yothers G, Goldberg R, de Gramont A (2009) Evidence for cure by adjuvant therapy in colon cancer: observations based on individual patient data from 20,898 patients on 18 randomized trials J Clin Oncol 27:872–877. doi: 10.1200/JCO.2008.19.5362 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Franke AJ, Parekh H, Starr JS, Tan SA, Iqbal A, George TJ Jr. (2018) Total Neoadjuvant Therapy: A Shifting Paradigm in Locally Advanced Rectal Cancer Management Clin Colorectal Cancer 17:1–12. doi: 10.1016/j.clcc.2017.06.008 [DOI] [PubMed] [Google Scholar]
- 5.Jessup JM, Goldberg RM, Asare EA, et al. (2017) Colon and Rectum. In: Amin MB (ed) AJCC Cancer Staging Manual, 8th.edn. Springer, Chicago (IL), pp. 251–274 [Google Scholar]
- 6.College of American Pathologists (CAP) protocol for the examination of resection specimens from patients with primary carcinoma of the colon and rectum. https://documents.cap.org/protocols/cp-gilower-colonrectum-resection-20-4100.pdf
- 7.Smith KD, Tan D, Das P, Chang GJ, Kattepogu K, Feig BW, Skibber JM, Rodriguez-Bigas MA (2010) Clinical significance of acellular mucin in rectal adenocarcinoma patients with a pathologic complete response to preoperative chemoradiation Ann Surg 251:261–264. doi: 10.1097/SLA.0b013e3181bdfc27 [DOI] [PubMed] [Google Scholar]
- 8.Wittekind C B JD, Compton C and Sobin L https://www.uicc.org/sites/main/files/atoms/files/TNM_FAQs_EN.pdf - TNM Supplement: A Commentary on Uniform Use Accessed February 17, 2021
- 9.Fortea-Sanchis C, Martinez-Ramos D, Escrig-Sos J (2018) The lymph node status as a prognostic factor in colon cancer: comparative population study of classifications using the logarithm of the ratio between metastatic and nonmetastatic nodes (LODDS) versus the pN-TNM classification and ganglion ratio systems BMC Cancer 18:1208. doi: 10.1186/s12885-018-5048-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Holleczek B, Rossi S, Domenic A, Innos K, Minicozzi P, Francisci S, Hackl M, Eisemann N, Brenner H, Group E-W (2015) On-going improvement and persistent differences in the survival for patients with colon and rectum cancer across Europe 1999–2007 - Results from the EUROCARE-5 study Eur J Cancer 51:2158–2168. doi: 10.1016/j.ejca.2015.07.024 [DOI] [PubMed] [Google Scholar]
- 11.NCCN (2021) National Comprehensive Cancer Network Esophageal and Esophagogastric Junction Cancers https://www.nccn.org/store/login/login.aspx?ReturnURL=https://www.nccn.org/professionals/physician_gls/pdf/esophageal.pdf, pp. [DOI] [PubMed]
- 12.Rullier A, Laurent C, Vendrely V, Le Bail B, Bioulac-Sage P, Rullier E (2005) Impact of colloid response on survival after preoperative radiotherapy in locally advanced rectal carcinoma Am J Surg Pathol 29:602–606. doi: 10.1097/01.pas.0000153120.80385.29 [DOI] [PubMed] [Google Scholar]
- 13.Shia J, McManus M, Guillem JG, Leibold T, Zhou Q, Tang LH, Riedel ER, Weiser MR, Paty PB, Temple LK, Nash G, Kolosov K, Minsky BD, Wong WD, Klimstra DS (2011) Significance of acellular mucin pools in rectal carcinoma after neoadjuvant chemoradiotherapy Am J Surg Pathol 35:127–134. doi: 10.1097/PAS.0b013e318200cf78 [DOI] [PubMed] [Google Scholar]
- 14.de Campos-Lobato LF, Dietz DW, Stocchi L, Vogel JD, Lavery IC, Goldblum JR, Skacel M, Pelley RJ, Kalady MF (2012) Clinical implications of acellular mucin pools in resected rectal cancer with pathological complete response to neoadjuvant chemoradiation Colorectal Dis 14:62–67. doi: 10.1111/j.1463-1318.2010.02532.x [DOI] [PubMed] [Google Scholar]
- 15.Karamchandani DM, Chetty R, King TS, Liu X, Westerhoff M, Yang Z, Yantiss RK, Driman DK (2020) Challenges with colorectal cancer staging: results of an international study Mod Pathol 33:153–163. doi: 10.1038/s41379-019-0344-3 [DOI] [PubMed] [Google Scholar]
- 16.Kirsch RMD SN, Dawson H, Driman DK (2019) Wide variability in assessment and reporting of colorectal cancer specimens among North American pathologists: results of a Canada-US survey Can J Pathol 11:59–69 [Google Scholar]
- 17.Foong KS, Mishra A, Guy R, Wang LM, Shepherd NA (2016) How do we stage acellular mucin in lymph nodes of colorectal cancer specimens without neo-adjuvant therapy? Histopathology 69:527–528. doi: 10.1111/his.12970 [DOI] [PubMed] [Google Scholar]
- 18.Gandhi J, Avery N, Strang J. (2015) Acellular mucinous lymph nodes in colonic mucinous adenocarcinoma International Archives of Medicine 8:1–3 [Google Scholar]
- 19.Cohen R, Buhard O, Cervera P, Hain E, Dumont S, Bardier A, Bachet JB, Gornet JM, Lopez-Trabada D, Dumont S, Kaci R, Bertheau P, Renaud F, Bibeau F, Parc Y, Vernerey D, Duval A, Svrcek M, Andre T (2017) Clinical and molecular characterisation of hereditary and sporadic metastatic colorectal cancers harbouring microsatellite instability/DNA mismatch repair deficiency Eur J Cancer 86:266–274. doi: 10.1016/j.ejca.2017.09.022 [DOI] [PubMed] [Google Scholar]
- 20.Popat S, Hubner R, Houlston RS (2005) Systematic review of microsatellite instability and colorectal cancer prognosis J Clin Oncol 23:609–618. doi: 10.1200/JCO.2005.01.086 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Figure 1: Kaplan-Meier curve comparing survival outcomes of all MMR-D tumors in three groups shows that the clinical outcome of patients with acellular mucin (cases; study group) was more similar to pN0 controls compared to pN+ (N1) colon cancer controls. The number of cases and events in this sub-analysis were too low to calculate p values.
Data Availability Statement
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.