Risk Factors and Screening Considerations

Factors associated with ASCC include HPV infection (predominately genotypes 16 and 18), human immunodeficiency virus (HIV) positivity, sexually transmitted infections, immunosuppression, and tobacco use.^{4, 6} HPV status and p16 overexpression correlate with survival and recurrence outcomes.^{7, 8} In HPV-positive tumors, the dysfunction of p53 due to the HPV-E6 protein can sensitize tumors to chemoradiation (CRT). In non-HPV mediated anal cancers (10%-15% of cases), p53 suppression is often unrecoverable due to gene mutations,⁷ lowering CRT effectiveness.^{9, 10} Non-HPV-16 genotypes are potentially more common among patients with HIV.¹¹

Individuals with a history of HPV-mediated gynecologic cancers are at high risk for ASCC. The International Anal Neoplasia Society’s (IANS) consensus guidelines recommend that women with a history of vulvar cancer or high-grade intraepithelial lesion (HSIL) to start screening for anal cancer within 1 year of diagnosis.¹² Screening for patients over 45 years old with a history of cervical or vaginal cancers or HSIL is determined on a case-by-case basis. A list of screening and diagnostic procedures is further reviewed in IANS consensus guidelines.¹² Furthermore, patients with a new diagnosis of anal intraepithelial lesion or ASCC are recommended to undergo screening for synchronous gynecologic malignancies or HSIL (cervical, vulvar, and/or vaginal) with a gynecologic examination including biopsy of suspicious lesions.¹³ A Swedish population-based study of more than 3.7 million women showed an association between history of grade 3 cervical intraepithelial lesions (CIN) and the risk of developing anogenital cancers.¹⁴ The risk of anal cancer was zero in the first year after a CIN diagnosis but increased yearly, with an incidence rate ratio of 4.98 after 10 years compared to women without a CIN diagnosis.¹⁴ Therefore, physicians should remain vigilant in screening for anal cancer in patients with a prior history of gynecologic cancers.

Pretreatment evaluation involves a complete history, physical examination, digital rectal examination (DRE), inguinal nodal evaluation and, if applicable, a gynecologic examination. Staging involves CT scans of the chest, abdomen, and pelvis; pelvic MRI aids in anatomy delineation, treatment planning, and evaluating suspicious findings.^{4, 13} F-18 fluorodeoxyglucose positron emission tomography/CT (FDG-PET/CT) is recommended for nodal staging and metabolic activity of suspicious features on CT and/or MRI.¹³

Standard-of-Care Management

Organ preservation is the standard of care for patients with nonmetastatic ASCC ( Figure 1 ). For localized tumors less than or equal to 2 cm (T1 per American Joint Committee on Cancer 8th edn), local excision with at least a 1 cm margin may be considered if anal function can be preserved.^{4, 13} Local excision should generally be reserved for patients with an anal margin or peri-anal tumors with no or minimal involvement of anal sphincter complex. Involved or close margins warrant repeat excision, although this is often challenging. For cases in which excision is not feasible, definitive CRT with 5-fluorouracil (5FU) and either mitomycin C (MMC) or platinum is preferred.

Patients with locoregional disease (T1-4 N0-1 M0) are generally recommended definitive MMC/5FU-based CRT. Radiation alone^{15 - 17} as well as MMC omission are associated¹⁸ with inferior disease control (with better toxicity profiles) compared with CRT. Although MMC remains standard, replacement with cisplatin may achieve similar disease control and decreased hematological toxicity.¹⁹ Capecitabine may be substituted for 5FU.^{20 - 22} There is no role for induction^{23, 24} or maintenance¹⁹ chemotherapy in nonmetastatic disease. Radiation doses are institution- and country-specific and range from 50 to 60 Gy to the primary tumor, 30.6 to 45 Gy to elective nodes, and 50 to 54 Gy to involved nodes.^{4, 13, 19, 24, 25} Intensity-modulated radiation therapy (IMRT) is associated with lower toxicity profiles and fewer treatment breaks compared with 3D conformal radiation therapy.^{22, 25 - 27} Proton radiation therapy has not been shown to improve disease control,²⁸ toxicity profile²⁹ despite favorable dosimetry, or patient-reported outcomes compared with photon-based radiation therapy.³⁰

Clinical response can continue up to 6 months post-CRT, even if a complete clinical response (cCR) is not observed by 3 months.³¹ Biopsy before 6 months post-CRT is not recommended. The primary method of assessing treatment response is DRE and anoscopy, typically 26 weeks after CRT in line with the ACT-II study.⁶ Post-treatment radiographic evaluation with pelvic MRI and/or FDG-PET/CT may also be utilized, although that is not routine. Surveillance for patients with cCR includes DRE and inguinal nodal examination every 3 to 6 months for 5 years, and anoscopy every 6 to 12 months for 3 years.³² In patients with stage II-III disease, imaging of the chest, abdomen, and pelvis is completed annually for 3 years.³² Approximately 80% of recurrences occur in the first 2 years post-CRT.³³ Up to a third of patients with persistent or recurrent disease will ultimately require salvage abdominal peritoneal resection or even pelvic exenteration depending on extent of disease.^{4, 13}

Patients with metastatic disease at diagnosis are recommended chemotherapy as first line with carboplatin + paclitaxel, cisplatin + 5 FU, or modified docetaxel + cisplatin + 5FU (mDCF).^{4, 13, 32, 34} The addition of checkpoint inhibition with chemotherapy as first-line treatment is institution dependent. While broadly speaking immunotherapy (IO) has been reserved as second-line treatment here, early readout from the PODIUM-303/InterAACT 2 study shows a modest PFS benefit (9.30 vs 7.39 months; P = 0.0006) with no difference in OS, although data is maturing.³⁵ Ongoing trials across the care continuum are summarized below. Five-year overall survival (OS) is 72% for all patients (76% for women; 64% for men), 86% and 39% for patients with localized and distant disease, respectively, with a trend for better outcomes among women.¹ Five-year overall and disease-free survival for those who underwent salvage surgery after CRT are lower, at approximately 40% to 50%.⁴

Figure 1.

Clinical management

There does not appear to be a difference in OS between HIV-positive and HIV-negative ASCC patients.^{36, 37} One series reported higher local failures among HIV-positive patients, which are likely due to toxicity-related treatment breaks.³⁸ Importantly, low CD4 count (< 350 cells in one study³⁹ and < 200 in others^{40, 41} ) and high viral load (> 700 copies/mL³⁹ ) correlated with increased grade 3 or higher toxicity, treatment interruptions, and hospitalizations. Wexler and colleagues reported that patients with low CD4 count and high viral load had significantly worse 5-year overall- and cancer-specific survival.³⁹ Collectively, those studies underscore the importance of treating HIV in this patient population. Additionally, a review of 13 population-based HIV and cancer registries throughout the US with 24,486 patients (10.9% with HIV and 9.3% with AIDS) showed that HIV was associated with increased all-cause mortality (1.53, 95% CI, 1.42-1.64) and with increased anal cancer-specific mortality among female patients (1.52, 95% CI, 1.18-1.97).⁴² The National Comprehensive Cancer Network provides a good review of the management of people living with HIV undergoing cancer treatment.⁴³

Tailored Treatment Strategies

Reducing long-term toxicity without compromising cancer control is the focus of ongoing trials. The PersonaLising RadioTherapy dOse for Anal Cancer (PLATO; ISRCTN88455282) integrates ACT-3, ACT-4, and ACT-5 to tailor management using biology and margin data ( Table 1 ).⁴⁴ For example, ACT-3 is a de-escalation phase 2 protocol that is evaluating observation in T1N0 patients with negative margins (no tumor on ink) post local excision while those with a close margin (≤ 1 mm) receive dose-reduced CRT (41.4 Gy/23 fractions).⁴⁴ Similarly, the ACT-4 study is evaluating dose-reduced CRT for patients with small tumors (T1-2N0 ≤4 cm). Early data show comparable 6-month cCR between dose-reduced (41.4 Gy/23 fractions) and standard (50.4 Gy/28 fractions) CRT, with lower toxicity in the dose-reduced arm.⁵³ These results support the feasibility of safely de-intensifying treatment in carefully selected patients. The ongoing DECREASE phase II trial⁵⁴ (ECOG-ACRIN 2182; NCT04166318) is evaluating de-escalated treatment for node-negative disease. T1N0 patients receive 36 Gy to the primary tumor and 32 Gy to elective nodes in 20 fractions, while T2N0 patients receive 41.4 Gy to the primary tumor and 34.5 Gy to elective nodes in 23 fractions. The standard arm delivers 50.4 Gy to the primary tumor and 42 Gy to elective nodes in 28 fractions. In addition, patients in the experimental arm receive a lower MMC dose (10 mg/m² vs 12 mg/m² ) and 1 less cycle of 5FU compared with the standard arm.

Conversely, treatment intensification is being explored for patients with more advanced disease, who are at higher risk of treatment failure. Secondary analysis of RTOG 9811 demonstrated poor 5-year disease-free survival (DFS) in patients with T3-4N+ disease ranging from 43% to 27%,⁵⁵ highlighting the need for more aggressive treatment. Within the PLATO framework, the ACT-5 trial intensifies radiation dose for high-risk patients (T3-4N0 or T2-4N+) up to 61.6 Gy in an effort to improve control. Moreover, a recent phase 3 study in Russia also evaluated adding paclitaxel to CRT in 144 ASCC patients (~72% N+; ~78% stage III).⁵⁶ The study was terminated prematurely in 2019 due to loss of access to mitomycin C. Within this limitation, paclitaxel appears to significantly improve 3-year DFS (87.1% vs 64.4%, P = .001) and OS (95.5% vs 80.0%, P < .001), with increased grade 3 to 4 toxicities (56.9% vs 26.4%, P < .0001), compared with CRT with doublet chemotherapy. This provides a signal for possible benefit of intensifying chemotherapy in this group of patients. Lastly, immunotherapy is being investigated in conjunction with chemotherapy for this higher risk population, which is reviewed later in this article.

While strides have been made in organ preservation and overall disease outcomes with CRT for patients with anal cancer, challenges remain in balancing long-term toxicity and treatment morbidity. There is a need to tailor these treatments to an individual patient’s anatomic stage (as in DECREASE and PLATO) and also their molecular signatures, which provides a richer overview of each tumor’s biology.

Emerging Biomarkers

Understanding the molecular interplay between HPV pathogenesis and genomic alterations is crucial for optimizing treatment outcomes and personalizing therapy. Patients with HPV-positive disease have better outcomes partly due to inherent HPV oncogenesis. PIK3CA mutations and PTEN loss are present in 30% and 14% of HPV-positive cases, respectively,⁷ which was also observed in an exploratory whole-exome sequencing (WES) analysis of RTOG 9811 patients ( n = 62).⁵⁷ In contrast, patients with HPV-independent disease are more likely to harbor higher p53 and CDKN2A mutation burdens, at 67% and 56%, respectively.⁷ In 2010, Lampejo and colleagues reviewed multiple biomarkers and reported a potential prognostic value for p21, Bcl-2, NF-kB, and cyclin A.⁵⁸

Exploratory analysis of RTOG 9811 noted additional mutations such as FBXW7 , which were prevalent in 15% of the cohort and associated with worse disease-free survival (hazard ratio [HR] 2.47 [1.02-5.96], P = .045) and a signal for inferior OS (HR 2.61 [0.97-7.04], P = .058).⁵⁷ Aldersley and colleagues performed WES on 72 patients with anal cancer ( n = 56 primary; n = 31 recurrent).¹¹ HPV integration was noted in 38% of cases and was more common in stages III-IV, at a rate of 2.69 integrations per sample compared with 0.91 integrations per sample for stages I-II ( P = .008).¹¹ They were numerically more common in recurrent and metastatic disease than in primary disease (1.88 vs 1.10; P = .092). The integration events were often associated with copy number variations and amplifications of genes such as PI3KCA , MYC , and CCND1 .¹¹ Interestingly, amplification of TERT and deletions of ATR , FANCD2 , and FHIT were reliably more common in recurrent/metastatic vs primary tumors, with corresponding enrichment of DNA damage response gene in recurrent tumors.¹¹ The authors posit that enrichment of those genes in the context of recurrent deletions may contribute to tumor recurrence post CRT. It is conceivable that HPV integrants across human cancers take advantage of host genomic aberrations, increasing instability, and ultimately leading to tumorigenesis early on and treatment resistance later.^{59, 60} Since viral integration and genomic instability may worsen as infected cells progress to malignancy,^{11, 59, 60} this provides a rationale for the prevention and aggressive treatment of premalignant lesions such as HSIL. The Anal Cancer-HSIL Outcomes Research (ANCHOR) study was a multi-institutional phase III study that sought to determine whether treating anal HSIL reduces the risk of progression to anal cancer among HIV-positive patients compared with active surveillance.⁶¹ Treatment included excision, ablation, or administration of topical agents. Active surveillance included high-resolution anoscopy at least every 6 months and an annual biopsy. In a cohort of roughly 4500 patients with a median follow-up of 25.8 months, risk of progression to invasive disease was reduced by 57% (95% CI, 6-80; P = .03) among those who received treatment.⁶¹

Liquid biopsies, including circulating tumor cells and circulating tumor DNA (ctDNA), have emerged as an adjunct in identifying genomic alterations and monitoring treatment response across various cancers, including gastrointestinal cancer. In the noncomparative phase II study SCARCE C17-02 PRODIGE 60, combining IO with chemotherapy, patients with complete molecular response (cMR) as measured by HPV ctDNA pre- and post-treatment had better 1-year progression-free survival (PFS) (60.4% vs 15.4%) and OS (90.7% vs 64.2%) compared with those without cMR, respectively.⁶² Moreover, Epitopes-HPV02 was a phase II single-arm study (NCT02402842) of patients with unresectable locally advanced/recurrent or metastatic ASCC where HPV ctDNA was evaluated as a predictive biomarker.⁶³ Positive HPV ctDNA at baseline did not correlate with PFS, although patients with a baseline ctDNA level < 2940 copies/mL had better PFS (HR, 2.1; 95% CI, 1.0-4.2; P = .04).⁶³ Like the SCARCE C17-02 study, cMR was associated with better 1-year OS (87% vs 50%) with an odds ratio of 7 (95% CI, 1.5-28.5; P = .02).⁶³ This is an exciting area of care in anal cancer; the use of blood biomarkers to guide clinical decision-making is under study in other virally⁶⁴ and nonvirally^{65 - 68} mediated cancers. While prospective studies are underway evaluating its role in guiding management,^{69, 70} data show that HPV ctDNA in anal cancer may indeed be an important prognosticator.

Furthermore, tumor-infiltrating lymphocytes (TILs) have been suggested to correlate with outcomes in patients with HPV-mediated disease, supporting a role for the way the adaptive immune system behaves in virally mediated cancers.^{9, 71} Patients with high TILs had significantly longer disease-free intervals compared with patients with absent/low TILs (92% vs 63 %; log-rank P = .006),⁷¹ in line with other HPV-mediated cancers.^{72 - 76}

The identification and integration of these biomarkers into clinical practice have expanded our understanding of ASCC, offering new avenues for targeted therapies. Immunotherapy has gained significant interest recently as a promising treatment for ASCC as it revolutionized the field of oncology with its success across many different cancers.

The Promise of Immunotherapy

HPV inherently furnishes an immunosuppressive and evasive environment through multiple mechanisms, one of which is upregulation of programmed death ligand-1 (PD-L1).^{77 - 79} While this provided a rationale to try different IO agents, outcomes in ASCC have been suboptimal.⁸⁰ Results of the randomized noncomparative phase II study SCARCE C17-02 PRODIGE 60 were recently published.⁶² In nonbiomarker selected, chemo-naïve, patients with locally advanced or metastatic ASCC, the addition of atezolizumab to mDCF vs mDCF alone did not meet the primary endpoint of 1-year PFS (45% vs 43%). The combination of mDCF + IO was associated with higher grades 3-4 (61% vs 42%) and serious adverse events (25% vs 12%).⁶² Interestingly, in patients with a PD-L1 combined positive score (CPS) of ≥ 5% ( n = 10), 1-year PFS with atezolizumab + mDCF was 70% (95% CI, 47-100) compared with 39% (24-62) in the CPS-negative group ( n = 28). This is in line with other studies showing IO responders are more likely to have higher PD-1/PD-L1 levels,^{81 - 83} albeit low response overall.

The INTERACT-ION is another phase II study from the French group that is studying the role of induction ezabenlimab, an anti-PD-1 antibody, in combination with mDCF as an induction regimen before CRT in treatment-naïve patients with locally advanced, stage III (T4N0 or TxN+), ASCC, with promising early results.⁴⁶ Moreover, dual checkpoint inhibition is of interest in ASCC as it has been shown to be more efficacious in activating the immune system.^{84 - 87} T cell immunoglobulin and ITIM domain (TIGIT) is an immune checkpoint receptor constitutively expressed on Tregs and is critical in mediating immunosuppression.⁸⁸ TIRANUS out of Spain is a single-arm phase II study (NCT05201612) that is studying the co-inhibition of PD-L1/TIGIT with atezolizumab and tiragolumab in combination with CRT for nonmetastatic patients with ASCC.⁴⁷

Several trials of IO are underway across the continuum of ASCC care. For instance, pembrolizumab is currently part of the single-arm phase Ib/II CoRInTH trial (NCT04046133) combining the PD-1 agent with CRT in locally advanced stage III-IV ASCC.⁴⁵ The National Cancer Institute (NCI) has 2 phase II studies with nivolumab following definitive CRT. The first study is in high-risk stage II-IIIB patients (EA2165; NCT03233711)⁴⁸ with primary endpoint of PFS. The second study is a risk-adapted trial (NCT04929028) of either nivolumab (high-risk, T3-4N0M0 or T2-4N1M0) or observation (low-risk, T1-2N0M0 or tumors <4 cm) following CRT in HIV-positive patients.⁴⁹ In more advanced disease, SPARTANA (NCT04894370) is a unique phase IIA study in metastatic ASCC that leverages radiation synergistic priming of the immune response (single-fraction 8 Gy to a target lesion) before starting spartalizumab (PD-1 inhibitor) and mDCF.⁵⁰ This regimen is then followed by consolidative ablative treatment to residual disease and maintenance spartalizumab, with the primary end point of PFS. Nivolumab (EA2176; NCT04444921)⁵¹ and retifanlimab (PODIUM-303/InterAACT 2; NCT04472429)⁵² are also being investigated in a phase III randomized fashion in combination with carboplatin/paclitaxel in metastatic and locally advanced/metastatic disease, respectively. As noted above, early data from the PODIUM-303/InterAACT 2 study show a signal of efficacy in terms of PFS but not OS at this time, with data continuing to mature. A summary of ongoing clinical trials is provided in Table 1 .

While studies are ongoing, IO holds great potential as an adjunct to standard-of-care management in all stages of the disease. However, nonbiomarker-driven IO studies may prove futile, underscoring the importance of personalizing therapy.