1 INTRODUCTION

Anal cancer is uncommon, yet it presents many challenges, including the need for permanent colostomy in most patients. According to the American Cancer Society, there have been 9440 new cases of anal cancer in the United States in 2022 and the number of new cases has been on the rise in the last few decades.¹

There are two main histologic types of anal cancer: squamous cell carcinoma (SCC) and adenocarcinoma. SCC accounts for more than 70% of anal cancers whereas adenocarcinoma represents approximately 5%–10% of cases. These incidences are in contrast to rectal cancer where adenocarcinomas are the most prevalent histology.^{2, 3} The exact origin of anal adenocarcinomas is not clear. The colorectal type of anal adenocarcinomas originates from the glandular cells of the transitional zone mucosa, whereas the extra-mucosal type originates from the anal canal glands and is usually associated with chronic untreated anal fistulas.^{3, 4}

According to the National Comprehensive Cancer Network (NCCN) guidelines, anal adenocarcinomas are managed with a similar strategy to rectal adenocarcinomas, with multi-modality treatment that includes neoadjuvant therapy followed by abdominoperineal resection (APR).⁵ This management strategy is different from the treatment of anal SCC, which involves sphincter-preserving radiation therapy with concurrent chemotherapy as the standard treatment, reserving APR for locally recurrent disease.⁶

A previous National Cancer Database (NCDB) study⁷ compared the treatment paradigms and outcomes of stage II–III anal SCC and adenocarcinomas. The study found that there were no significant differences in overall survival (OS) of patients with anal adenocarcinomas according to the treatment type, whereas treatment of SCC with chemoradiation alone was associated with 33% lower mortality. However, the previous analysis did not match the treatment groups of anal adenocarcinoma for demographics and disease stage. Therefore, we conducted the present study to assess the impact of neoadjuvant therapy on the outcomes of APR in patients with stage II–III anal adenocarcinomas after matching for baseline confounders using the propensity-score matched method.

2 PATIENTS AND METHODS

3 RESULTS

3.1 Description of the entire cohort

After screening the records of 5594 patients with anal adenocarcinoma, 742 patients met the study criteria and were included (Figure 1). A total of 417 (56%) patients were male and the mean age was 63.6 ± 12.4 years. The majority of (84%) patients were white while 11.6% were Black and 2.8% were of Asian origin. A total of 22.4% of patients had a Charlson score ≥1. The majority (97.4%) of patients lived in metropolitan or urban areas. A total of 50% of patients had stage II disease and 50% had stage III disease. A total of 81.4% of adenocarcinomas were not-specified, whereas 16.6% were mucinous adenocarcinomas and 2% were signet-ring cell carcinomas. A total of 27.5% of adenocarcinomas were of high grade. The 5-year OS rate was 56.2% and the median follow-up was 51.4 months (IQR: 29–81.2).

3.4 Pathologic outcomes and survival

Patients who received neoadjuvant therapy had more pT0 tumors (15.7% vs. 0%), fewer pT3-4 tumors (28.4% vs. 36.4%, p = 0.001), fewer pN1-2 tumors (22.9% vs. 34.7%, p < 0.001), more pTNM 0–1 disease (24.1% vs. 9.6%, p = 0.021), and less lymphovascular invasion (16.2% vs. 40%, p < 0.001). Neoadjuvant therapy was associated with significantly fewer harvested lymph nodes (median: 11.5 vs. 14, p = 0.009) and fewer involved lymph nodes. The two groups had similar rates of positive surgical margins, high-grade adenocarcinomas, and mucinous tumors (Table 3).

Table 3. Pathologic outcomes of patients who received or did not receive neoadjuvant therapy.

		Before matching			After matching
Factor	Group	Neoadjuvant (n = 533)	No neoadjuvant (n = 189)	p value	Neoadjuvant (n = 214)	No neoadjuvant (n = 107)	p value
Pathologic T (%)	0	64 (13.9)	0 (0.0)	<0.001	32 (15.7)	0 (0.0)	0.001
	1	69 (15.0)	12 (7.7)		29 (14.2)	11 (11.1)
	2	149 (32.4)	82 (52.9)		73 (35.8)	48 (48.5)
	3	107 (23.3)	32 (20.6)		40 (19.6)	21 (21.2)
	4	37 (8.0)	19 (12.3)		18 (8.8)	15 (15.2)
	IS	2 (0.4)	0 (0.0)		0 (0.0)	0 (0.0)
	X	32 (7.0)	10 (6.5)		12 (5.9)	4 (4.0)
Pathologic N (%)	0	307 (67.2)	99 (64.7)	0.207	143 (71.1)	59 (60.2)	0.102
	1	85 (18.6)	31 (20.3)		33 (16.4)	20 (20.4)
	2	32 (7.0)	16 (10.5)		13 (6.5)	14 (14.3)
	X	33 (7.2)	7 (4.6)		12 (6.0)	5 (5.1)
Pathologic TNM (%)	0	15 (4.1)	0 (0.0)	<0.001	10 (6.0)	0 (0.0)	0.021
	I	61 (16.6)	10 (6.9)		30 (18.1)	9 (9.6)
	II	147 (39.9)	81 (56.2)		65 (39.2)	46 (48.9)
	III	134 (36.4)	51 (35.4)		57 (34.3)	37 (39.4)
	IV	11 (3.0)	2 (1.4)		4 (2.4)	2 (2.1)
Lymphovascular invasion (%)	No	313 (84.1)	99 (63.1)	<0.001	124 (83.8)	54 (60.0)	<0.001
	Yes	59 (15.9)	58 (36.9)		24 (16.2)	36 (40.0)
Surgical margins (%)	Negative	483 (90.4)	157 (85.8)	0.096	180 (87.8)	85 (82.5)	0.225
	Positive	51 (9.6)	26 (14.2)		25 (12.2)	18 (17.5)
Median number of examined nodes [IQR]		12.00 [6.00, 16.00]	14.00 [9.00, 19.00]	<0.001	11.50 [6.00, 16.00]	14.00 [9.00, 19.00]	0.009
Median number of positive nodes [IQR]		0.00 [0.00, 1.00]	0.00 [0.00, 1.00]	0.107	0.00 [0.00, 0.00]	0.00 [0.00, 2.00]	0.002

• Note: Bold text in p value columns indicates statistical significance.
• Abbreviation: IQR, interquartile range.

Subgroup analysis of the pathologic outcomes according to tumor histology revealed that neoadjuvant radiation was associated with significant downstaging of the T stage in both non-mucinous (p = 0.017) and mucinous/signet ring cell carcinomas (p = 0.008). The rates of complete mucosal response (pT0) were similar in the two groups (15.6% and 16.7%, respectively). Neoadjuvant radiation was associated with a lower incidence of lymphovascular invasion and fewer harvested lymph nodes in both groups (Table 4).

Table 4. Pathologic outcomes of matched patients who received or did not receive neoadjuvant therapy according to tumor histology.

Factor	Group	Non-mucinous adenocarcinoma			Mucinous and signet-ring cell carcinoma
		Neoadjuvant	No neoadjuvant	p value	Neoadjuvant	No neoadjuvant	p value
Number		177	84		37	23
Lymphovascular invasion (%)	Yes	19 (15.7)	26 (37.7)	0.001	5 (19.2)	10 (47.6)	0.059
	No	102 (84.3)	43 (62.3)		21 (80.8)	11 (52.4)
Pathologic T stage (%)	0	26 (15.6)	0 (0.0)	0.017	6 (16.7)	0 (0.0)	0.008
	1	24 (14.4)	9 (12)		4 (11.1)	2 (8.7)
	2	59 (35.3)	32 (42.7)		14 (38.9)	15 (65.2)
	3	29 (17.4)	19 (25.3)		11 (30.6)	2 (8.7)
	4	18 (10.8)	12 (16)		0 (0.0)	3 (13.0)
	X	11 (6.6)	3 (4)		1 (2.8)	1 (4.3)
Pathologic N stage (%)	0	117 (71.3)	46 (61.3)	0.294	25 (69.4)	13 (56.5)	0.474
	1	27 (16.5)	15 (20.0)		6 (16.7)	5 (21.7)
	2	11 (6.7)	10 (13.3)		2 (5.6)	4 (17.4)
	X	9 (5.5)	4 (5.3)		3 (8.3)	1 (4.3)
Pathologic TNM (%)	0	8 (5.8)	0 (0.0)	0.051	2 (7.1)	0 (0.0)	0.596
	1	26 (19.0)	7 (9.7)		3 (10.7)	2 (9.1)
	2	51 (37.2)	36 (50.0)		14 (50.0)	10 (45.5)
	3	49 (35.8)	27 (37.5)		8 (28.6)	10 (45.5)
	4	3 (2.2)	2 (2.8)		1 (3.6)	0 (0.0)
Surgical margins (%)	Negative	151 (89.9)	67 (82.7)	0.15	29 (80.6)	18 (81.8)	1
	Positive	17 (10.1)	14 (17.3)		7 (19.4)	4 (18.2)
Median number of examined lymph nodes [IQR]		11.00 [6.00, 17.00]	14.00 [8.00, 18.75]	0.053	12.00 [8.00, 15.00]	16.00 [12.50, 19.00]	0.031
Median number of positive lymph nodes [IQR]		0.00 [0.00, 0.00]	0.00 [0.00, 2.00]	0.011	0.00 [0.00, 0.00]	0.00 [0.00, 2.00]	0.111

• Note: Bold text in p value columns indicates statistical significance.
• Abbreviation: IQR, interquartile range.

The 5-year OS rate of patients who received or did not receive neoadjuvant therapy was similar (52.1% vs. 50%, p = 0.812). The mean OS was statistically similar between the two groups (47.5 vs. 44.8 months, p = 0.253) (Figure 2). Cox regression analysis showed that the omission of neoadjuvant therapy was not associated with a significant increase in the hazards of mortality (hazard ratio = 1.195, 95% CI: 0.857–1.667, p = 0.294).

4 DISCUSSION

In the present study, although neoadjuvant therapy managed to downstage anal adenocarcinomas and achieve favorable pathologic outcomes; OS was similar to patients who did not receive neoadjuvant therapy. Neoadjuvant radiation did not have a significant impact on short-term clinical outcomes, except for lower 90-day mortality.

In the unmatched cohort, neoadjuvant radiation was administered more often to younger and healthier patients, who would predictably tolerate the adverse effects of neoadjuvant treatment better than elderly patients with comorbidities who are assumed to experience more G3/4 toxicities of preoperative chemoradiotherapy.¹⁰ Also, patients with more advanced disease were more likely to have neoadjuvant therapy, however, patients with poorly differentiated carcinomas received less neoadjuvant treatment, probably due to the lower response rate these tumors have to neoadjuvant therapy than moderately- and well-differentiated tumors.¹¹

After matching for possible confounders that may introduce selection bias, neoadjuvant therapy was associated with lower pathologic T and N staging than the group that did not receive neoadjuvant treatment. Moreover, approximately 16% of patients who had neoadjuvant therapy achieved a pathologic complete mucosal response, within the range of complete response of rectal cancer to neoadjuvant therapy of 15%–27%.^{12, 13} Similar to patients with rectal cancer, those with anal carcinoma who achieve a complete response may be eligible for watch and wait strategy with close surveillance¹⁴ to avoid non-restorative proctectomy and terminal colostomy that negatively impacts the quality of life.¹⁵ Future prospective trials are needed to assess the rates of clinical complete response after neoadjuvant radiation for anal adenocarcinoma that may enable organ preservation. However, it should be noted that while complete mucosal response (pT0) can be accurately assessed using MRI, endoscopy, and biopsy, complete nodal response (pN0) may not be as accurately assessed given the poor concordance between radiologic and pathologic N stage.¹⁶

Furthermore, neoadjuvant radiation was able to reduce the incidence of lymphovascular invasion, a negative prognosticator in colorectal cancer,¹⁷ to 16% compared to 40% when neoadjuvant therapy was not given. Du¹⁸ and colleagues also found neoadjuvant radiation to reduce lymphovascular invasion in rectal cancer by 5%, although this reduction was not statistically significant, probably because the authors used neoadjuvant radiation therapy only, and perhaps the addition of concurrent chemotherapy may enhance the therapeutic effect of radiation on cancer cells involved in the vessels.

Despite the positive downstaging effect of neoadjuvant therapy in achieving favorable pathologic outcomes, it did not prolong OS compared to patients who did not have neoadjuvant therapy. This finding may seem counter-intuitive to the studies that reported improved OS with neoadjuvant therapy for rectal adenocarcinoma.^{5, 19} However, anal carcinomas are different from rectal carcinomas in many aspects, including tumor biology and lymphatic pathways. A recent NCDB analysis of the trends of anal adenocarcinomas showed that the standard treatment of anal adenocarcinoma with the best survival outcome is surgery preceded by neoadjuvant chemoradiation.²⁰ The lack of a survival benefit of neoadjuvant therapy in our study might be explained by the effect of adjuvant chemotherapy that was given to 41.1% of patients who did not have neoadjuvant therapy versus 24.8% of patients who received neoadjuvant treatment. As also shown in the previous NCDB analysis,²⁰ giving adjuvant therapy after surgery can prolong OS by 6 months compared to surgery alone. This disparity of adjuvant treatment may explain the similar OS since patients who did not receive neoadjuvant therapy and were found to have stage III disease after surgery had adjuvant therapy, whereas patients who were down staged after neoadjuvant therapy did not require adjuvant therapy. Ultimately, both groups had similar OS, indicating that the survival benefit of radiation therapy in stage II–III anal adenocarcinomas is similar, whether given before or after surgery. This finding may help clinicians tailor therapy to each patient, depending on several factors related to the patient's condition and available resources.

Neoadjuvant therapy was associated with a significant reduction in the odds of 90-day mortality in our study. The explanation of this interesting finding is not clear and may be attributed to the downstaging effect of neoadjuvant therapy that may have facilitated radical resection, unlike advanced T4/N1-2 anal carcinomas that were operated on without neoadjuvant treatment, and thus resection might have been demanding and complicated. It is noteworthy that neoadjuvant radiation may be followed by an increased rate of perineal complications and surgical site infections after APR.^{21, 22} However, neoadjuvant therapy is not significantly associated with higher 30-day or 90-day mortality after rectal cancer surgery.²³

Despite the benefits of neoadjuvant therapy, it was associated with a longer period from diagnosis to surgery and fewer harvested lymph nodes than the no-neoadjuvant group. The time from diagnosis to surgery was almost tripled when neoadjuvant therapy was given. However, the prolonged wait before surgery may be justified by the positive downstaging effect of therapy that was able to achieve a complete response in 16% of patients. The fewer number of examined lymph nodes after neoadjuvant therapy has been previously reported in studies on rectal adenocarcinoma. However, as Billakanti et al.²⁴ concluded, although neoadjuvant chemoradiation decreased lymph node harvest in rectal cancer surgery, this reduced nodal harvest does not reduce survival.

It is important to note that, although neoadjuvant radiation therapy was associated with significant downstaging of both mucinous and non-mucinous anal adenocarcinomas, this pathologic benefit did not translate into a tangible survival benefit. This finding emphasizes the importance of selecting patients with stage II/III anal adenocarcinoma for neoadjuvant radiation rather than using it for all patients to avoid unnecessary delay in surgical management and potential postradiation perineal wound complications.

The current analysis adds to the current literature as most previous studies included smaller cohorts or did not match the compared groups for important confounders. The present study is one of the first large studies to assess the benefits of neoadjuvant therapy in anal adenocarcinoma, using propensity-score matching to minimize selection bias. The conclusions of the study, that while neoadjuvant radiation was associated with significant downstaging of anal adenocarcinoma but not a significant survival benefit and that approximately 16% of patients had a complete mucosal response to therapy that may render them candidates for the Watch and Wait strategy, may help inform the current practice and direct future research. However, there are limitations to the study that include its retrospective nature and lack of data on the chemoradiation regimen used and disease-free survival. Stratifying OS by disease stage and patients' demographics was not possible because of the relatively small number of patients included after matching.

5 CONCLUSIONS

Neoadjuvant therapy before APR was associated with significant downstaging of anal adenocarcinomas, reduction of the incidence of lymphovascular invasion, and lower 90-day mortality. However, OS was not improved when neoadjuvant therapy was given before surgery.

CONFLICT OF INTEREST STATEMENT

Dr. Wexner reports receiving consulting fees from ARC/Corvus, Astellas, Baxter, Becton Dickinson, GI Supply, ICON Language Services, Intuitive Surgical, Leading BioSciences, Livsmed, Medtronic, Olympus Surgical, Stryker, Takeda and receiving royalties from Intuitive Surgical and Karl Storz Endoscopy America Inc. Dr. Emile reports receiving consulting fees from SafeHeal.

SYNOPSIS

This study aimed to assess the impact of neoadjuvant therapy on the clinical and pathologic outcomes and overall survival (OS) of patients with stage II–III anal adenocarcinomas after abdominoperineal resection (APR). Neoadjuvant therapy before APR was associated with significant downstaging of anal adenocarcinomas and lower 90-day mortality, yet similar OS to patients who were surgically treated without neoadjuvant treatment. Both groups had similar conversion rates, hospital stay, 30-day mortality, 30-day readmission, and positive surgical margins.