肛門癌風險：以人類乳突狀瘤病毒HPV為基礎之子宮頸篩檢計畫

資料來源：刺胳針感染症醫學期刊，www.thelancet.com/infection Vol 19 August 2019，

財團法人台灣紅絲帶基金會編譯

與子宮頸癌相似，大多數肛門癌（89-100％）由高危型別之人類乳突狀瘤病毒（HPV）持續感染所引起，尤其是HPV16型。肛門癌僅佔所有下消化道惡性腫瘤的4％，但其發病率在高所得國家中呈現上升之趨勢。對於美國而言，報導之每年平均增加的百分比變化，在男性為2.1而女性則為2.9。在美國每年約有8,300例新發之肛門癌病例被診斷出來，女性的發病率高於男性，在歐洲亦有類似的數字被估計。

超過三分之一的肛門癌患者在診斷後5年內死亡。肛門癌已確定之幾個風險群體，例如男男性接觸者、移植接受者，愛滋病毒/愛滋病感染者，以及有HPV感染引起之婦女子宮頸癌、陰道癌或外陰癌病史的患者，而後者之與一般人群相比，其肛門癌的發生率明顯增加（達3至22倍）。

由HPV誘導的肛門與生殖器癌症，在很大程度上是可以預防的，對於尚未感染HPV的人(HPV-naïve) 可透過預防接種（一級預防）或篩檢癌前病變（二級預防）。對於子宮頸癌，有些國家基於人群的細胞學篩查，已被透過初次HPV篩查或共同檢測（HPV篩檢和細胞學檢查）所取代。相較之下，肛門癌篩查僅建議用於高風險群體，如愛滋病毒陽性者，尤其是男男間性行為者。然而，利益上之證據尚有待持續的前瞻研究來建立。

在「刺胳針感染症」上的一項研究中，Chunqing Lin及其同事，已經證明從常規HPV子宮頸篩檢計畫中獲得的數據，可用於定義參與婦女的肛門癌風險輪廓。林和他的同事做了回顧性合作匯總薈萃分析，包括來自36項研究13, 427名婦女個人層級的數據，在這些婦女中配對的肛門和子宮頸樣本均可獲得。他們發現子宮頸HPV16與肛門HPV16間有密切相關。 41％的HIV陰性有子宮頸HPV16感染的女性也有肛門HPV16感染，相較於沒有子宮頸HPV16者則只有2％（盛行比[PR] 16.5）；而愛滋病毒陽性女性，這些值分別為46％和11％（PR 4.4）。

無論HIV感染狀態如何，肛門癌前病變（高度鱗狀上皮內病變，high-grade squamous intraepithelial lesion , HSIL）與子宮頸高危之HPV和子宮頸之高度鱗狀上皮內病變有關。四分之一有子宮頸HPV16的老年女性（年齡> 45歲）有肛門HPV16相關的高度鱗狀上皮內病變（分別在25％的愛滋病毒陰性女性，和23％的愛滋病毒陽性婦女中）。

儘管文獻中所報導由肛門的高度鱗狀上皮內病變到侵襲性癌症的進展率各有不同，Lin及其同事的研究結果具有重要公共衛生意涵。可能建議所有感染子宮頸高危型別HPV之患者進行肛門癌篩檢（不論其愛滋病毒感染狀況如何），尤其如果檢測到的HPV型別是HPV16更應如此。此外在HIV陰性的女性中，有子宮頸高度鱗狀上皮內病變或癌症的一項診斷似乎是肛門高度鱗狀上皮內病變上很強的決定因素（PR 23·1）。在HIV陽性的女性中，則發現子宮頸高度鱗狀上皮內病變患者中有25％亦有肛門高度鱗狀上皮內病變，但子宮頸細胞學檢查正常者中也有7％的患者有肛門高度鱗狀上皮內病變。同樣的，8％沒有子宮頸高危型別HPV感染的HIV陽性婦女有肛門高度鱗狀上皮內病變。這個發現可能會導致愛滋病毒陽性婦女應進行更多肛門癌篩檢之爭辯，而不只是針對那些與HPV相關的不典型增生的人。在HIV陰性和陽性婦女兩者間的肛門癌症篩檢算法上，若納入其他風險因素可能會進一步更完善，諸如吸菸或性行為。

和肛門癌相類似的與HPV相關之口咽癌其發生率正在穩步增加，尤其是在男性，而女性亦有關。在分析子宮頸有關之生物指標時，如前述Lin及其同事描述的（HPV16陽性，高度鱗狀上皮內病變HSIL，和癌症）等等，是否也與HPV誘導的口咽癌風險之增加有關，將是非常的有趣。這樣的研究可以在有全面性的健康數據庫和癌症登記之國家進行。但是，可行的HPV相關之口咽癌的篩查試驗尚不存在。高分辨率的肛門鏡檢查就是診斷肛門癌前病變的黃金標準，但高分辨率肛門鏡檢查缺乏廣泛的可用性，也阻礙了除HIV陽性的個體之外其他高風險群體的全面篩查。HPV疫苗接種已經導致年輕女性子宮頸高危型別HPV感染和癌前病變等比率上的重大變化。子宮頸（和肛門）HPV16 / 18的感染和癌前病變之減少，儘管在資源不夠情況下，讓針對繼發的肛門癌之預防目標更為可行，特別是如果肛門高危型別HPV之檢測可以在高分辨率的肛門鏡檢查之前進行。更期待的是，Lin和他的同事所提出的研究結果能鼓勵前瞻性試驗的開展，以便去評估肛門癌風險升高的所有女性之篩查計畫。

Anal cancer risk: HPV-based cervical screening programmes

Similar to cervical cancer, most anal cancers (89–100%) are induced by persistent infections with high-risk human papillomavirus (HPV), especially HPV16.  Anal cancer accounts for only 4% of all malignancies of the lower alimentary tract, but its incidence is rising in highincome countries. For the USA, increasing average annual percentage changes of 2.1 for men and 2.9 for women have been reported.  About 8,300 new anal cancer cases are diagnosed in the USA each year, with higher incidence rates in women than in men.  Similar numbers have been estimated for Europe.  More than one-third of patients with anal cancer die within 5 years after diagnosis.  Several risk groups for anal cancer have been identified, such as men who have sex with men, transplant recipients, people with HIV/AIDS, and women with a history of HPV-induced cervical, vaginal, or vulvar cancer. The latter group has an increased incidence (three to 22 times) of anal cancer, compared with the general population.

Anogenital HPV-induced cancers are largely preventable by prophylactic vaccination of HPV-naïve individuals (primary prevention) or by screening for precancers (secondary prevention).  For cervical cancer, population-based cytology screening has been replaced by primary HPV screening or by co-testing (HPV and cytology) in several countries.  By contrast, anal cancer screening has been recommended only for high-risk groups, such as HIV-positive individuals, especially men who have sex with men.  However, the evidence of benefit has yet to be established in ongoing prospective studies.

In a study in The Lancet Infectious Diseases, Chunqing Lin and colleagues have shown that data

available from routine HPV-based cervical screening programmes can be used to define anal cancer risk

profiles in the participating women. Lin and colleagues did a retrospective collaborative pooled meta-analysis, including individual-level data from 36 studies with 13,427 women for whom paired anal and cervical samples were available. They found that cervical HPV16 is strongly associated with anal HPV16. 41% of HIVnegative women with cervical HPV16 also had anal HPV16, compared with 2% of those without cervical HPV16 (prevalence ratio [PR] 16·5); in HIV-positive women, these values were 46% and 11% (PR 4·4). Anal precancer (high-grade squamous intraepithelial lesion [HSIL]) was associated with cervical high-risk HPV and with cervical HSIL, regardless of HIV status. One quarter of older women (aged >45 years) with cervical HPV16 had anal HPV16-associated HSIL (25% of HIV-negative women, 23% of HIV-positive women).

Notwithstanding the divergent progression rates of anal HSIL to invasive cancer reported in the literature, the findings of Lin and colleagues have important public health implications. Anal cancer screening might be advisable in all women with cervical high-risk HPV infection (irrespective of their HIV status), particularly if the detected HPV-type is HPV16. Furthermore, a diagnosis of cervical HSIL or cancer seems to be a strong determinant for anal HSIL, also in HIV-negative women (PR 23·1). In HIV-positive women, anal HSIL was found in 25% of patients with cervical HSIL, but also in 7% of those with normal cervical cytology. Similarly, 8% of HIV-positive women without cervical high-risk HPV infection had anal HSIL. This finding could argue for anal cancer screening in more HIV-positive women than only in those with HPV-associated dysplasia.  Anal cancer screening algorithms for both HIV-negative and HIV-positive women could possibly be further refined by incorporating additional risk factors, such as smoking or sexual practices.

 Similar to anal cancer, incidence rates of HPV associated oropharyngeal cancer are steadily increasing, especially in men, but also in women. It would be interesting to analyse whether the cervical markers described by Lin and colleagues (HPV16 positivity, HSIL, and cancer) are also associated with an increased risk for HPV-induced  oropharyngeal cancer. Such studies could be done in countries that maintain comprehensive health databases and cancer registries. However, viable screening tests for HPV-associated oropharyngeal cancer do not yet exist. High-resolution anoscopy is the gold standard for diagnosing anal precancer, but highresolution anoscopy is not widely available, which could hamper comprehensive screening of high-risk groups beyond HIV-positive individuals. HPV vaccination has already led to significant changes in cervical highrisk HPV infection and precancer rates in younger women. The decrease of cervical (and anal) HPV16/18 infection and precancer could make targeted secondary anal cancer prevention feasible despite insufficient resources, especially if high-resolution anoscopy could be preceded by anal high-risk HPV testing. Hopefully, the findings presented by Lin and colleagues will encourage the initiation of prospective trials evaluating screening programmes for all women with an elevated risk for anal cancer.

www.thelancet.com/infection Vol 19 August 2019