The screening program
The late diagnosis of lung cancer is the main reason to why the disease has the highest death toll out of all cancer types, causing 1,69 million deaths in 2015 . This is due to the fact that the symptoms of lung cancer often occur in a later stage of the disease. At that point, the cancer can be hard to treat and only 5-14% of the patients survive the first 5 years. If lung cancer could be detected earlier, many lives could be saved. One way to efficiently detect lung cancer in people would be to introduce a screening program .
The Swedish National board of health and welfare takes the decision whether a screening program is needed. To do so, the board has 15 criteria that are taken into consideration. These criteria are built on the government organisation Socialstyrelsen’s criteria for screening and are also mentioned in Cancerfondsrapporten 2016 (Cancer Fund report of 2016) that was made available to the team by Aina Törnblom at The Swedish Cancer Society, Cancerfonden [3,4].
The main points comprises that:
- The disease should be common, severe and cause strain on the society
- Efficient treatment should be available
- Early detection and treatment should lead to a higher survival
- The diagnostic test should with high probability classify truly diseased people as diseased and truly healthy people as healthy
- Diagnosis of symptom free individuals should be possible
- The function should be commonly accepted by the population that is included in the screening and it should be cost effective
Being the fifth most common cancer form but still having one of the worst prognosis, as well as costing the swedish society 3,6 billion SEK in 2013 , the first criteria must be said to be met. Availability of efficient treatments and the fact that earlier detection and treatment should lead to higher survival rate can be demonstrated by the estimation from Cancerfonden, that about 200 to 500 people, of patients dying each year of lung cancer in Sweden could be saved with a screening program .
During the development of a screening program for lung cancer, including the system and prototype created by the team, the first step was to investigate the screening programs currently ongoing. The screening program for breast cancer and cervical cancer were the initial inspiration. In the design of the screening program the following factors from the criterias has been taken into consideration: Frequency, patient group, economical model and examination method. To make the screening program efficient, high participation is of outmost importance. With this in mind, a survey was created to ensure the interest of the public and to be able to include the potential participant’s wishes and thoughts in the design of the screening program.
The frequency of the examinations were discussed from a cost and time consumption perspective but also determined by the speed of the disease progression, although the progression of lung cancer progression can vary greatly. According to Pedersen et al. (2017), the optimal screening interval is not yet determined. However, a suggestion for the Nordic countries is a baseline screening, followed by a single annual examination that results in a biennial examination in participants without pulmonary nodules, where pulmonary nodule is defined as round or oval growths in the lung. One suggestion based on the article as well as the frequency of breast cancer examinations, is to have a lung cancer examination every two years .
The distributed survey regarding public opinion of a screening program for lung cancer, show that down to one year recurrent examinations was not considered a problem, and as many as 89% would have attended if they were called.
The question of who to include in the screening program was discussed with two parties from different areas; Tommy Bjork, chairman of the patient organization Stödet and Dr. Hirsch Koyi, Associate professor at the Department of Respiratory Medicine at Gävle University that is working on bringing forth a model for implementing a screening program in the nordic countries.
According to Cancerfonden, the largest part of the patients suffering from lung cancer are people over 60 years old that have been smoking for a longer time period . The initial thought was therefore that the screening program would include the part of the population above 60 years of age. After discussions with Mr. Björk and Dr. Koyi, this idea was challenged. Mr. Björk thought that the best would be to only include smokers since it otherwise would make the program too complex and expensive. This is supported by the fact that approximately 90% of all lung cancer patients have smoked at some point in their life. This cutoff was also suggested by Dr. Koyi in a skype meeting that was held during the summer where he proposed the idea to only include people that have been smoking for 30 years. It is also supported by the American Cancer Society's recommendation that people that have been smoking a package of cigarettes a day for 30 years should be included in a screening program .
Inclusion of only smokers in the program was thoroughly discussed especially from a practical and ethical point of view. Questions that came up included how the smokers would be documented and how long you have to smoke to be counted as a smoker? Would it be ethically correct to document the population’s smoking habits? From another perspective; would it be fair to exclude the few but still existing non-smokers that are in fact also discovered to suffer from lung cancer?
In the end, it was concluded that the most efficient method would be to only screen smokers that have smoked for more than 30 years of their life. This would reduce the cost and required organization for the program and since as many as 5 200 patients die from lung cancer every year due to smoking, it was considered a legitimate cutoff . The documentation can be done in several ways but Cancerfonden suggests that the primary health care system could be involved to reach the people in the risk group and refer to the screening program. This could be considered too risky and causing many people in the risk zone to slip through the cracks. Another idea is to send out forms to the part of the population that turn 60 where they have a chance to report their smoking habits. In this way it is voluntary to give out the information. This is a part that has to be continuously worked with to assure that as many people in the risk zone as possible are reached and that people’s privacy is ensured.
The sampling would occur at a hospital by trained healthcare personnel using the BREATHtaking device , see Figure 1. Before the examination, the patient will be instructed not to eat a few hours before the examination, which was recommended by Dr. Koyi. This because different type of foods can secrete VOCs and therefore interfere with the results. When beginning the examination, the patient will be instructed to inhale and then blow into the mouthpiece of the device which will lead the air through a filter collecting the VOCs present in the breath. This will be repeated five times to make sure the VOCs are enough concentrated. The examination will take a few minutes and when it is finished the filter containing the VOCs will be sent to a laboratory facility for analysis.
When the filter arrives to the laboratory, the VOCs will be released from the filter and incubated with the yeast biosensor. The yeast will then be plated and incubated in 30°C for a few days to study the possible red colour change. The patient will be informed of the result and if the yeast had no colour change, the patient will be continuously be called for screening examinations every two years. If the yeast cells would turn red and thereby indicate that the patient has lung cancer, further examinations will be needed. The patient would therefore do a CT scan to examine the position and size of the tumor to determine continued treatment.
To ensure that the method is suitable for a screening program, the goal with the biosensor is to have a high detection rate, so as not to overlook any patients with lung cancer, as well as a low number of false positives. However, a number of experiments and statistical analyses will be needed to assure this. Currently, the biosensor is in the “proof of concept”-stage and therefore it would be hard to fulfill the screening condition “the diagnostic test should with high probability classify truly diseased people as diseased and truly healthy people as healthy” today, even if it aims to fulfill the condition in the future.
When instituting a lung cancer screening program, cost efficiency is of great importance. Medical personnel costs will be roughly the same regardless of which method that is used. However, use of the biosensor in a screening program would result in a reduction of the cost compared to the use of a CT scan where a single examination costs 15 000 to 20 000 SEK (around 1 900-2 500 USD) . Besides medical personnel, the use of the biosensor would result in additional costs confined to laboratory personnel and culturing of the yeast which is regarded relatively cheap. For the device itself, the filter would be considered the highest cost, with around 4 300 SEK (around 530 USD) per filter .
The survey brought up economical aspects to include the general public’s view on how the program should be financed and how much they themselves would be willing to pay for an examination. The survey was divided into smokers and non-smokers since it was considered interesting to study the difference in perspective. For smokers as many as 95,5% would attend a free of charge, annual examination if they were called to it. If the examination instead would cost each participant money, the participation was reduced to 69,7%. For non-smokers the free of charge examination would have a participation of 87,8% that was reduced to 54,5% when charged. This shows that smokers generally are more inclined to test themselves for lung cancer and that more smokers are willing to pay for an examination. This is also shown by the results from the question of how much the participants would be willing to pay. For the smokers, 4,5% would be willing to pay more than 900 SEK (~115 USD, see Figure 2) but for the non-smokers, the figure was instead only 0,8% , see Figure 3.
As seen in the survey, the interest for paying for lung cancer screenings is higher among smokers compared to the rest of the population. According to Mr. Björk there is a mentality around lung cancer that smokers only have themselves to blame if they get the disease. This mentality can lead to objections if the screening program were to be funded by the government. But early detection could save the society money especially in combination with quit-smoking campaigns.
To conclude, the first three points in the criteria for a screening program proposed by Socialstyrelsen, have been shown to be fulfilled. This is further supported by the article by Detterbeck et al. (2013), that confirms the need for a screening program for lung cancer and what benefits it would give . Due to the early stage of the biosensor it is hard to ensure compliance of the two criteria with respect to the diagnostic device. This is regarded as fundamental properties when developing the sensor further and will always be kept in mind. The last criteria is considered fulfilled, both due to the cost efficiency of the biosensor which was established earlier as well as based on the results from the survey. It showed a high willingness to participate in a future screening program for lung cancer.
From the beginning of the project, the idea to create a prototype that could be used for a potential screening program seemed very inviting. By creating such a device, it would also be necessary to truly consider the potential use of the project. The features found most important in a good final product were safety and ease of use, low rate of misdiagnosis as well as low price and thus availability to many in need.
To get started, a prototype brainstorming meeting enabled different ideas to be discussed and the best alternatives are presented in this paragraph. One conclusion made, was that the accuracy of the test is less dependent on the prototype construction than on the system design, which can be read about in the Project, description. This part of the wiki focuses on how to make use of the system in the best possible way.
The first question faced in the brainstorming was whether the sensor should be possible to use at home or if it should only be used in a controlled environment such as a hospital. The initial idea was to have a yeast filter which one could exhale through, exposing the yeast to potential VOCs. After careful consideration this alternative was deemed too riskful. For people without proper knowledge, handling genetically modified yeast might pose too much of a risk, especially since the yeast would be close to the mouth. Therefore it would be better if the test was handled at a hospital by personnel with appropriate education. One problem with this idea could be that doctors or nurses generally are not used to working with yeast and thus extra personnel would be needed. Also, with the yeast on a filter it would still be close to the mouth. Since it was decided that simplicity and safety are important characteristics of the device, the conclusion of the initial thoughts was that it would be preferable if doctors or nurses not were required to handle the yeast and that the yeast would not be located on a filter. With this in mind, a new idea took form.
The sample bags
Commercially available sample bags that could be used to collect and store VOCs from the exhaled breath were found after some research. If such bags were used, the risk for patients and doctors to come in contact with yeast would be eliminated. After filling the bags with exhaled air they would be sent to laboratory facilities where incubation with the yeast would be conducted. This idea appeared safer and more practical, but new questions arose. How would the yeast be exposed to the VOCs? How much air would be needed to give enough VOCs for the yeast to be able to sense it?
The concentration of VOCs in the breath is low, in the parts per billion range for both octanal  and butanone . The sensitivity of GPCRs is varying. According to Minic et. al  Ri7 can sense 5*10-14 - 5*10-4 M while Olfr1258 can sense around 5*10-4 M . Since the amount of VOCs needed is unsure, it is safe to assume that a couple of liters of air would most likely be needed. However, to keep the yeast incubated with that much air requires a large, bulky vessel. This could be solved by creating a recirculation system with the air from the sample bags bubbling through the suspended yeast culture. In that way the yeast could be exposed to all the collected air but only a smaller amount at a time.
Consulting with experts
Meeting with Dr. Carl Johan Franzén
To discuss the design of such a recirculation system, Dr. Carl Johan Franzén was contacted. Dr. Carl Johan is Associate Professor in Bioreaction engineering at Chalmers and has many years of experience in working with yeast. The general idea considered was to have a tube with a specially designed lid that would be compatible with the recirculation system. With the use of a compressor, the VOCs could be pumped through the yeast media. Air from the tube should also be led back to the sample bag through a pipe. With enough recirculation and a small enough vessel the advantage would be that a shaking incubator would not be needed. The discussion also covered whether the system would have a problem with oxygen limitations. Dr. Carl Johan gave the advise to calculate how much oxygen that is needed and compare to how much breath contains. If the amount of oxygen in the breath is not enough, it might be possible to dilute the collected breath with oxygen. This solution would require the sensitivity of the receptors to be taken into account to ensure high enough VOC concentration to match the range of recognition for the receptors. A third option would be to switch between the collected air, containing VOCs and pure oxygen. To minimize the oxygen consumption the growth rate of the cells should be kept low, possibly with limiting media.
Dr. Carl Johan also thought that the use of a spectrophotometer or a plate reader should be considered since it would allow the amount of cells that turn red to be measured. If it would be possible, the plating of the cells would be avoided. It would however be more expensive. An advantage of using a plate is the ease of analysis since it does not require additional instrumentation.
Consulting with Prof. Per Sunnerhagen
To learn more about the mating process and the ADE2 gene and how it affects the colour of the cells, an email conversation with Prof. Per Sunnerhagen at University of Gothenburg was initiated.
He explained that he did not believe oxygen limitation to be a problem since the mating process is not particularly oxygen demanding. Additionally, he mentioned the possibility to see the difference in colour of the cells, in liquid media, if the cell density is high enough. He was not aware of the ability of a spectrophotometer to detect the red colour but gave the advice that ADE2 mutants accumulates a polymer of aminoimidazoleribotide which generates a red-orange fluorescent that can easily be seen in a microscope. That ADE2 mutant fluorescence could be helpful both for designing the prototype but also in the lab, where it might be difficult to distinguish the red colour in a colony if only a few of the cells are stained.
Contact with PExA
In addition to consulting experts, the Swedish market was researched since all inputs are beneficial when developing a suitable screening strategy and prototype for the project. The goal was to find companies and professionals working within the field that could aid in modifying and improving the idea. One company that the iGEM team came in contact with was PExA (Particles in Exhaled Air). The company works with sampling small particles from exhaled air.
Founded in 2014, PExA is a company that have created a new method to collect particles from the lungs. The company divides the method into two big parts, sampling and analysis of particles. The sampling device is non-invasive and easy to use. The patient exhales repeatedly into a mouthpiece connected to the machine and the exhaled particles are then collected on a membrane. Each sample usually consists of ~20% proteins and ~80% phospholipids, which are used for analysis. The machine is today involved in different research projects for example the search for new biomarkers for early detection of lung diseases.
Since they are experts in the area of breath analysis and sampling correlated to sickness, a contact was established with PExA and a study visit was scheduled. The aim of the meeting was to learn more about the technique, the market and to get feedback on the iGEM project.
At the meeting the team started with a short introduction about the project and the representative from PExA, Svante Höjer, in charge of the development at the company, held a short presentation of the company. The main part of the meeting focused on discussing the project.
PExA explained how they collect the particles with a Teflon membrane which makes it possible and easy to concentrate the particles at one place. This fact made the team rethink the original sampling idea, which comprised the use of airbags designed for containing VOCs. One of the problems with the airbags is the need for a large amount of air to obtain a high enough concentration of VOCs. This is problematic both due to the constraint put on the patient as well as the inefficiency in the transportation of large volumes of air. Also, the air can only be stored in the bags a limited amount of time, or the VOCs can condense on the walls of the bag. If possible, it would therefore be preferable to instead use a small fiber.
Designing the prototype
After the new idea had arisen, a brainstorming session was held to research the possibilities to use a fiber and to discuss which idea to further work with as well as their advantages and disadvantage.
The conclusion from this session was that the idea with the fiber was the one to be developed. SPME and similar fibers absorbing VOCs are commercially available and could be used for our system. The largest advantage with this idea is that the device would be more manageable and that the fiber would be much easier to transport from a hospital to the laboratory facilities. A potential drawback though would be the extra step needed to extract the VOCs from the fiber in order to expose the yeast to them. Though compared to the recirculation system in the sample bags idea, the extraction step is probably less cumbersome. Available SPME fibers could possibly be used as they are, but a more optimized version would be to prefer.
One idea would be to use a porous fiber as a filter, place this in the device and exhale through it. Important in such a design would be that the fiber-filter is not too compact which would make exhaling through the device difficult. The fiber could be sent to a laboratory facility where the VOCs is extracted and added to the yeast media.
For the design of the prototype, a relatively small device was considered desirable in order to make it easy to use. It would also be good, for both practical and environmental reasons, if as much as possible of the device was reusable to enable easy cleaning. A two-part option was therefore considered. One main part where the fiber-filter could be placed and one mouthpiece to blow through. In this design, only the filter would be disposable.
Another important part of the design is to minimize the impact of the surroundings. Since the system will be very sensitive, it is important to avoid that air, other than the exhaled air from the patient, comes in contact with the fiber. For this purpose, valves in the mouthpiece and in the end of the device could be used. To control the background levels of the VOCs, the test should take place in a controlled hospital or laboration facility with oxygen masks or VOCs free air to inhale if necessary. Since the secretion of VOCs also can depend on the food intake, the patient is restricted not to eat a couple of hours before the test, in order for the result to be reliable .
With these thoughts in mind the prototype was designed in CAD, as may be seen in Figure 4. In addition to this, the team also contacted a former iGEM team member David Hansson, now working with his master’s thesis on the department of Systems and Synthetic biology at Chalmers University of Technology. He offered to aid with 3D-printing of the prototype and he printed the first prototype after the provided design.
After some consideration the prototype was deemed slightly too big. The mouthpiece could be too big and therefore hard to blow through and the whole device seemed unnecessary long. David made it shorter and scaled it down and printed it again. This time the prototype was perceived too small for holding in the hand even though the mouthpiece was more suitable. The design was therefore settled on a mixture of these two alternatives. The mouthpiece from the smaller version was extended but the size was otherwise kept. The main part had the same size as in the bigger version but was shortened. The final version of the prototype is shown in Figure 5.
The idea of how to use the device is to place the fiber-filter in the opening, exhale through the device, pause for a minute, exhale through again and repeat this a few times. After that the fiber-filter can be sent to the laboratory facilities where the VOCs are extracted and added to the yeast. The color of the yeast cells are then closely watch and analysed, hoping that the yeast don’t turn red.
-  World health organization Media centre. Cancer fact sheet. 2017. Available from: http://www.who.int/mediacentre/factsheets/fs297/en/. Accessed: 2017-09-17.
-  American Cancer Society, Can Lung Cancer Be Found Early?. Available from: https://www.cancer.org/cancer/lung-cancer/prevention-and-early-detection/early-detection.html 2016. Accessed: 2017-09-17.
-  Socialstyrelsen. Nationella Screeningprogram, Modell för bedömning, införande och uppföljning. 2014. Available from: http://www.socialstyrelsen.se/Lists/Artikelkatalog/Attachments/19360/2014-2-16.pdf. Accessed: 2017-09-17.
-  Cancerfonden. Cancerfondsrapporten 2016. 2016. Available from: https://www.vardfokus.se/globalassets/2016/april/cancerfondsrapporten-2016.pd.pdf. Accessed: 2017-09-01.
-  Lundqvist A, Andersson E, Steen Carlsson K. Kostnader för cancer sverige idag och år 2040. Institutet för Hälso- och Sjukvårdsekonomi; 2016. IHE-rapport; 2016:1.
-  Pedersen, J. H., Sørensen, J. B., Saghir, Z., Fløtten, Ø., Brustugun, O. T., Ashraf, H., ... & Nyrén, S. (2017). Implementation of lung cancer CT screening in the Nordic countries. Acta Oncologica, 1-9.
-  Cancerfonden: Hedefalk B. Lungcancer [Internet]. Cancerfonden; 2015 [Updated 2015-01-28; cited 2017-09-26]. Available from: https://www.cancerfonden.se/om-cancer/lungcancer
-  Cancerfonden: Biörnstad L. Rökning [Internet]. Cancerfonden; 2016 [Updated 2016-05-20; cited 2017-09-26]. Available from: https://www.cancerfonden.se/livsstil/rokning
-  Cancerfonden. PET-röntgen hittar tumören [Internet]. 2009 [Cited: 2017-09-24]. Available from https://www.cancerfonden.se/artiklar/pet-rontgen-hittar-tumoren.
-  SPME fiber assembly Carboxen/Polydimethylsiloxane (CAR/PDMS) sheet, price and Availability. 2017. Available from: http://www.sigmaaldrich.com/catalog/product/supelco/57318?lang=en®ion=SE
-  Detterbeck, F. C., Mazzone, P. J., Naidich, D. P., & Bach, P. B. (2013). Screening for lung cancer: diagnosis and management of lung cancer: American College of Chest Physicians evidence-based clinical practice guidelines. CHEST Journal, 143(5_suppl), e78S-e92S.
-  Fuchs P, Loeseken C, Schubert J and Miekisch W. (2009). Breath gas aldehydes as biomarkers of lung cancer. International journal of cancer. Journal international du cancer, 126. 2663-70. 10.1002/ijc.24970.
-  Fu X-A, Li M, Knipp R J, Nantz M H and Bousamra M. Noninvasive detection of lung cancer using exhaled breath. Cancer Medicine 2014; 3(1): 174–181
-  Minic J, Persuy M-A, Godel E, Aioun J, Connerton I, Salesse R, and Pajot-Augy E. Functional expression of olfactory receptors in yeast and development of a bioassay for odorant screening. FEBS Journal, 272(2):524-537, 2005
-  Suzuki Y and Shimono K. Deciphering the receptor repertoire encoding specific odorants by time-lapse single-cell array cytometry. Scientific Reports, 6(19934):1-9, 2015