编者按：由中国抗癌协会主办，天津医科大学肿瘤医院、天津市抗癌协会、中国整合医学发展战略研究院承办的“2023中国整合肿瘤学大会（2023 CCHIO）”将于2023年11月16—19日在天津举办。国际乳腺疾病协会（SIS）常务理事Alexander Mundinger博士将在本次大会上带来题为乳腺癌筛查成本效益（Cost-effectiveness of breast cancer screening）的演讲。《肿瘤瞭望》在大会前夕采访了Mundinger博士，分享更多有关乳腺癌筛查策略、成本效益及人工智能应用等问题的看法。

 

The 2023 Chinese Congress of Holistic Integrative Oncology(2023 CCHIO)will be held in Tianjin from November 16 to 19，2023.Dr.Alexander Mundinger，who is on board of directors of the International Society of Breast Diseases(SIS)，will deliver a speech(Cost-effectiveness of breast cancer screening)at this congress.He was interviewed by Oncology Frontier on the eve of 2023 CCHIO.

 

Mundinger博士：根据最新统计数据，比利时的患病率和发病率最高，不过欧洲也有一些低发病率地区。亚洲的发病率也较低。一项针对中国农村地区乳腺癌筛查的研究表明，农村地区的乳腺癌患病率只有城市的一半。然而，这些农村地区的乳腺癌死亡率要远高于城市地区。纵观全球，情况亦是如此。我们甚至无法明确一些极度贫困地区的发病率情况，就目前了解的情况，死亡率相当高，确诊时间也很晚。

 

所以我认为这个问题的关注点是，只用一种筛查方法，如乳腺钼靶，是否能满足不同国家的筛查需求。对于这个问题，世界卫生组织和相关机构给出的答案非常明确。如果财政资源不足或者有其他优先事项，那么这个国家就无法把所有资源都倾注于乳腺癌的检测和治疗。因此，在这些极度贫困的国家，对症状明显的癌症进行诊断和治疗才是当务之急，这需要癌症专家、专科治疗中心的支持，以及提供不同于西方国家的诊断筛查方法。在西方国家，乳腺钼靶是首选方法，而触诊和超声可能更适合这些国家。西方国家多采用空心针活检（CNB）或负压真空辅助抽吸活检（VAB），而贫困国家更适合使用细针穿刺活检（FNA），因其成本低。

 

另外，在任何情况下，触诊降低晚期癌症死亡率都是一件极其困难的事情。有许多研究和大型试验表明触诊筛查并不能降低死亡率，因此应关注无症状阶段的乳腺癌治疗，包括手术和药物治疗。这种思路背后的要点是是症状性癌症降期。西方国家的肿瘤治疗效果非常好，主要由于大多数患者为1期肿瘤，非常小的肿瘤，死亡率非常低，甚至与普通人群相当。因此，死亡率与癌症分期和大小关系密切。

 

乳腺钼靶筛查在日常中更常使用，超声、数字乳腺摄影术（DBT）、增强钼靶和磁共振成像（MRI）这些现代筛查方法都有不错的筛查效能，但都比较昂贵。我们无法在全球癌症发病率和患病率不等（GlobalCan的数据）的地区内采用相同的方法进行筛查。

 

Oncology Frontier:What are the differences in prevalence of breast cancer globally?How might it affect breast cancer screening in different places?

 

Dr.Mundinger:The prevalence is very different.Maybe the highest one，according to new statistics，are in Belgium，the highest prevalence and incidence，and we have low incidence areas in Europe.And also in Asia，there are lower incidences.There is a study of screening in China rural areas.The prevalence is half of the prevalence in the urban regions.On the other hand，the mortality is much higher in these rural areas compared to the urban areas.And the same is true for the whole world.We have very poor areas.And we are not sure whether we know precisely the prevalence.But we know the mortality is very high.The cancers are detected lately.

 

And so your question focuses on，should we treat all the countries with the same method?For example，mammography screening.And the answer given from WHO and others is very clear.If you don’t have any financial resources for example，and you have other priorities in your country，you cannot spend everything for detecting and treating breast cancer.So the idea is focusing in these very poor countries on symptomatic cancers and to make the diagnosis and the treatment of these symptomatic cancers better.We know how to make it better.You need cancer specialist，specialized centers.And you need another approach.In contrast to the Western countries where mammography is the number one，palpation and ultrasound are probably more suitable for these countries.And then if you are always adhering to core biopsy in the Western countries or vacuum-assisted biopsies，probably for those poor countries，FNA approach is more suitable because it’s less expensive.

 

But on the other hand，it’s very difficult to lower the mortality of symptomatic extended cancers，whatever you do.Because we have multiple studies，big trials that show that palpation is not able to reduce the mortality.So the focus should probably be the treatment，the surgical and the pharmaceutical treatment of these cancers after diagnosis of a symptomatic stage.And the idea behind is to downstage these symptomatic cancers.So it’s a different approach.In the Western countries，we have excellent results.So we are biased because we are hunting very small cancers.And in the stage one，the mortality does not exist.We have the same mortality compared to the normal average society.So mortality depends on the stage and on the size of the cancer.

 

And with modern mammography screening compared with an assessment with ultrasound and other modern tools we are using every day，like DBT，or contrast-enhanced mammography for assessment，and MRI，we are able to have excellent results.But these methods are all expensive.And we cannot use the same methods for the whole spectrum of prevalences and incidences that is given by GlobalCan，for example.

 

Mundinger博士：在这个问题上首先需要清楚的是成本-效益的原理和定义。进行筛查会有一定的成本，但是这不代表不进行筛查就会更省钱，我们要做的，是将进行筛查的成本控制在政府能接受的范围之内，即他们愿意为此投入的资金门槛。通常情况下，这个资金门槛介于人均GDP的一到三倍不等，这对于研究的可比性也是有一定意义的。

 

在两项比较中国农村地区乳腺癌高低风险患者的研究中，该资金门槛为人均GDP的三倍。而在大多数欧洲国家，通常将其控制在人均GDP的一倍以下。为了保持标准化，我建议采取与中国研究相同的方式，将其设定为人均GDP的三倍。否则，我们难以科学地比较阈值介于一到三倍间不等的各种情况。

 

不过最后还是得由政府决定。比如拉丁美洲的某些城市犯罪率很高，每天被枪击的人数比乳腺癌诊断的人数还要多。另外贫困、婴儿死亡率高，处于战争状态的国家，他们有其他各种问题需要优先考虑，也就无暇顾及乳腺癌筛查。

 

因此，必须结合当地的情况，并在一个筛查系统中进行适当整合。我们的使命是为患有癌症的妇女提供帮助，这是我们的初心和使命，所以无法顾全所有情况也无可厚非。

 

Oncology Frontier:What do you think is cost-effective for breast cancer screening in different regions?

 

Dr.Mundinger:So you have to be familiar with the principles and definitions of cost-effectiveness.And it does not mean it is less expensive than no screening.It means it is less expensive than the threshold that needs to be defined by，let me say，the politicians.And this threshold is a willingness to pay.And usually，it’s between one to three times the gross domestic product per person or per capita.And for comparability in studies，it maybe make sense.

 

It was the case also in the two big Chinese studies in rural areas and comparing high-risk，low-risk patients.It was three times the GDP per capita.And in most European countries，it’s possible just to be below one times the GDP per capita.So for standardization，I would suggest，as the Chinese studies did it three times the GDP per capita or per person.Otherwise，you don’t know if you try to compare one to three times.It’s very difficult to compare it from a scientific point of view.

 

But finally，the politicians have to make up the decision.And if a country is suffering from a high rate of criminality in some cities in Latin America，more people are shot per day than breast cancer diagnoses are made.There are other priorities.If poverty，if a high rate of babies dying is a problem，you have to focus on these issues.And another point is nations in war，they have completely other facets than take care of breast cancer screening.

 

So the local situation has to be just considered and to be wisely integrated in such a system.But our mission is to argue for the women who have cancer.And therefore，we cannot consider all the other conditions.We have to focus on our mission.

 

Mundinger博士：这个方向潜力无限。已经有学者对这方面展开了研究，虽然一开始的研究设计可能无法尽善尽美。但也有诸如MASAI试验的研究，明确了将AI整合进行现今的筛查系统是可行的，主要有以下三种潜在可行的方法。

 

第一种是一个独立的AI筛查系统，虽然可能不是目前最佳的方法，但是可行的。第二种是一个辅助放射科医师的支持系统，可以充当第二阅片员来帮助筛查。这个方式在当下是合理且有效的。第三种方法是在两个医师阅片后，运用AI来排除显示出癌症可能性极低的乳腺X线照片。以上是AI乳腺癌筛查的三种应用场景。

 

我们知道人工智能是一个出色的工具，在筛查方面会发挥很大的作用，因为它可以在保证质量的前提下降低成本。或许在一个筛查系统中，如果医师的培训不够充分，人工智能说不定可以帮助提高筛查质量，但这也仅限于那些具有数字化基础设施和大量财力的国家。另外，中国是世界上发表了最多人工智能相关研究的国家之一，也是高度数字化的国家。

 

Oncology Frontier:How do you think of the application of artificial intelligence in breast cancer screening?

 

Dr.Mundinger:It’s a great future horizon.And it has already begun.And we have，in the beginning，not so well-designed studies.But in the meantime，we have several studies like the MASAI study that clearly shows it’s possible to integrate systems.And there are three facets that are possible.

 

The first facet is just a standalone system.This is probably not currently the best approach，but it works.The second approach is just to have a support system that assists the radiologists.And such a support system also can be a second reader.This is a rational approach that will be possible with a modern system.And the third approach is just to delete those clients on mammography with a very low probability to show any cancer from the two readers.So these are the three indications.

 

And we have to consider that AI is a fantastic tool.And it will help in screening a lot because it can reduce the costs.And finally，it will not decrease the quality.Maybe it will increase the quality if in a system are not so well-trained readers.But it’s a privilege for some countries with a digitalized structure with a lot of money.And by the way，China is a country that has the most publications of artificial intelligence and is highly digitalized.