Catching cancer in the act

A new generation of tests is attempting to identify the disease much earlier, assess treatment effectiveness and assist in the choice of therapy

Multiple test: electrochemical cell used to simultaneously detect three proteins linked to prostate cancer

Léo RamosMultiple test: electrochemical cell used to simultaneously detect three proteins linked to prostate cancerLéo Ramos

Brazilian researchers working at home and abroad are completing work on a new generation of laboratory tests for earlier cancer detection, before it becomes identifiable through clinical examinations. Generally speaking, the earlier the disease is discovered, the greater the chance of successful treatment and even a cure. There are at least four new tests. Developed by teams in São Paulo, São Carlos and Spain, they use different strategies to capture signs of tumors in samples of blood, urine and other body fluids. If the tests prove efficient in the next assessment stages through which they have yet to pass, they may be able, in specific situations, to replace more invasive tests such as biopsies and punctures and serve as a complement to clinical examinations and imaging.

The two tests at the most advanced stage of development were designed by a team of geneticists led by Anamaria Camargo at the Ludwig Institute for Cancer Research (LICR) and the Molecular Oncology Center of Hospital Sírio-Libanês (HSL), in São Paulo. Both are based on the genetic analysis of the tumor characteristics of each patient and represent a step toward personalized medicine, which makes it possible to perform more accurate diagnoses and prescribe treatment tailored to each patient. “Customization is taking place in all areas of medicine, but it is more developed in oncology, because of the genetic basis of cancer,” says Camargo. In Brazil, this model, which depends on identifying the genetic cause of diseases (see story), is still in its infancy. The model is beginning to be implemented in some private hospitals and is now gaining momentum in São Paulo State with a combined initiative of five research centers working on a project to advance personalized medicine (see article).

The first test consists of analyzing a gene panel to guide cancer therapy. To prepare the panel, the researchers cross-checked information on recurrent genetic alterations in human tumors with information about the molecular pathways altered in these tumors and drug targets commonly used to treat the disease. In all, they selected 494 genes that were the most frequently altered in different tumors and that served as the target for a certain type of medication. “This panel can help guide treatment, because some of these mutations cause the tumor to be sensitive to certain compounds,” says Camargo.

Advanced cancer treatment centers, such as Memorial Sloan Kettering Cancer Center in New York, or MD Anderson Cancer Center in Texas, also have their own panels, each with a different number of genes. And they already make them available so that their doctors can select the most effective drug to treat each patient.

Slides for samples of tumor DNA to be sequenced at the Molecular Oncology Center of Hospital Sírio-Libanês (HSL)

Léo RamosSlides for samples of tumor DNA to be sequenced at the Molecular Oncology Center of Hospital Sírio-Libanês (HSL)Léo Ramos

Camargo and her group are currently working to validate the gene panel at HSL. They sequenced the genetic material from the tumors of 12 people treated at the hospital and identified the specific mutations of each cancer. “We have already analyzed seven cases and we are completing another five,” says Camargo. So far, she says, the test produced at the hospital showed 100% sensitivity and 100% specificity and did as well as larger panels numbering 600 genes, developed by Foundation Medicine, an American genomic analysis company, and available in the market.

In another evaluation, Luis Felipe Campesato, a biomedical doctor and Peter Galante, a bioinformatic specialist, both of HSL, demonstrated that the hospital panel can also be used to guide cancer treatment with a class of drugs that became available only in recent years: immunotherapeutics.

They are molecules that stimulate the immune system to attack tumor cells and they are producing promising results against some types of tumors, in particular skin and lung tumors. “Treatment with these drugs is expensive and only benefits some patients,” says Camargo. “So it is important to identify who will respond.”

Using a bioinformatics strategy, Campesato and Galante compared the capacity of the hospital’s panel and the Medicine Foundation’s panel to associate the number of skin and lung tumor mutations (mutational load) with the response to immunotherapy. In a study published in October 2015 in the journal Oncotarget, they demonstrated that both were as efficient as sequencing the entire human genome.

According to the study, immunotherapy was effective in 70% of patients with lung cancer who had a high number of genetic alterations. In successful cases, patients remained cancer free for at least six months after treatment—half of them presented no signs of tumor 18 months after the use of the medication. Yet among patients with few gene alterations, the compounds worked in only 20% of the cases.

“Our test proved to be feasible from a scientific standpoint, but now we have to demonstrate its practical application,” says Luiz Fernando Lima Reis, a biochemist and HSL’s head of research. Before the test can be used by the hospital’s physicians, it must go through a certification phase and be scaled up.

In parallel to developing the gene panel, Camargo’s group is working on an individualized test to determine if the antitumor treatment is doing what it is supposed to do, and to detect any recurrence of the disease. Her team began working on this test at the request of a group of gastrointestinal surgeons, Angelita Habr-Gama and Rodrigo Oliva Perez of the Angelita e Joaquim Gama Institute.

In the early 1990s, Habr-Gama, a researcher and internationally respected surgeon, proposed a bold and less aggressive strategy to treat certain cases of rectal cancer. The standard therapy to treat rectal tumors was removal of the end portion of the intestine, followed by a treatment based on radiation therapy and chemotherapy to prevent tumor recurrence. In searching for a way to avoid the intestinal removal, she reversed the order of the therapy and began to treat her patients first with radiation and drugs, followed by close monitoring with imaging. She managed to avoid surgery in 28% of her cases (see Pesquisa FAPESP Issues nº 162 and nº195).

Preparation of the test for prostate cancer: capturing protein biomarkers in blood serum samples using magnetic particles

Léo RamosPreparation of the test for prostate cancer: capturing protein biomarkers in blood serum samples using magnetic particlesLéo Ramos

Given the risk of the cancer recurring, Habr-Gama and Perez partnered with Camargo to produce a molecular test to detect as soon as possible if the treatment worked or if the problem recurred. They devised a customized genetic test that has now been proven feasible, but needs improvements.

Starting from a sample of the tumor, the researchers determine the patient’s specific genetic alterations. Upon completion of the combined treatment, they then collect only blood samples from time to time to measure the amount of circulating tumor DNA. In principle, this strategy should be able to find residual tumors after therapy, in addition to metastatic foci not detectable by clinical examination or imaging.

A pilot test of four people with rectal cancer showed the test was still not as sensitive as the researchers initially hoped in terms of detecting residual disease after therapy. After treatment, it was unable to detect the so-called minimal residual disease, in which less than 10% of the initial number of tumor cells remains. But it was able to detect the presence of cancer when the figure surpassed 20%. The test, however, proved to be very effective in evaluating the response to the treatment and in detecting a reduction in tumor size. With the test, they were able to follow progression of the disease and detect metastases clinically within a period of 18 months.

This result, also published in the October 2015 issue of Oncotarget, suggests that the test can potentially serve as a response marker to the treatment and monitoring of the disease after therapeutic intervention. Tumor DNA levels in the blood decreased in cases where the combination therapy of radiation and drugs worked. They increased in two patients where the tumor began to grow again.

“The data suggest that the amount of altered genetic material circulating in the blood is proportional to the size of the tumor,” says Lima Reis. The hospital team now plans to use the test with 20 more patients at Sírio-Libanês.

While the São Paulo researchers are looking for ways to detect tumors from their genetic material diluted in blood, another researcher in São Carlos, Ronaldo Censi Faria, is working to improve the accuracy of the blood test used to make the initial diagnosis of prostate tumors. Faria is a chemist and professor in the Chemistry Department of the Federal University of São Carlos (UFSCar), and he has developed a sensor to simultaneously identify three blood proteins associated with prostate cancer.

Today the blood test used to identify changes in the prostate only measures the level of prostate specific antigen (PSA), which may indicate the presence of tumors before clinical symptoms of cancer. The problem is that in some cases of the disease the PSA has not changed, while in other cases where its level appears elevated there is no tumor is present. “So we work with the detection of PSA-specific membrane antigen or PSMA prostate and platelet factor 4, PF-4, for a more accurate diagnosis,” says Faria. “The idea is to reduce the risk of false results.”

Faria began developing the sensor between 2011 and 2012, during his postdoctoral studies at the University of Connecticut in the United States. The detection of biomarkers occurs through the emission of light, the result of a chemical reaction (electrochemiluminescence). The intensity of light in the device is proportional to the concentration of proteins in the blood. The sensor contains disposable graphite electrodes on which antibodies are deposited. When one of the three biomarkers interacts with the antibodies, a chemical reaction occurs and produces light.

Sensor developed at UFSCar: disposable graphite electrodes with deposits of antibodies against cancer biomarkers

Léo RamosSensor developed at UFSCar: disposable graphite electrodes with deposits of antibodies against cancer biomarkersLéo Ramos

Another type of device was developed by a Brazilian chemist, Priscilla Monteiro Kosaka, working at the Microelectronics Institute of Madrid (IMM), in Spain. Called a nanomechanical sensor, it is made ​​of silicon and shaped like the diving board of a swimming pool, as a base and a “plank,” whose size is only half a millimeter. “Each diving board vibrates at a specific resonance frequency,” says Kosaka. “But when something is deposited on its surface, this resonance frequency changes in proportion to the mass of the deposit.” If the sample of blood contains a cancer biomarker, two changes occur in the sensor: its resonance frequency changes and the sample changes color.

Both Faria and Kosaka maintain that their devices are more sensitive and accurate than the existing diagnostic methods. “In the system we have developed, the detection thresholds are up to a thousand times lower than ELISA tests, in concentrations on the order of femtogram per milliliter,” says Faria. “This makes possible a large dilution of human serum, which leads to less sample consumption and minimizes potential interference.”

In addition, Faria says that the sensor he developed simultaneously detects more than one biomarker. “The number of false positives and negatives is approximately 40% using only PSA as an indicator of prostate cancer,” he says. “Multiple detection using three different proteins will allow a more accurate diagnosis.”

Kosaka also says that her nanossensor is very sensitive and specific.

“In laboratory tests simulating a blood sample, there were two errors per 10,000 tests,” she says. “Our method is able to identify the biomarker even if its concentration is less than 100 molecules in a blood sample.”

Due to the high levels of sensitivity and specificity of these devices and their low cost, they are likely to be used in routine blood tests, according to Faria. “This could impact public health because the chances of a cure, in the case of cancer, are higher when the diagnosis is made ​​early, and [that’s when] treatment costs are much lower,” he says.

The bad news is that this can not wait. “It can still take up to 10 years to reach the market,” says Kosaka. “Our goal is an ultrasensitive and inexpensive nanosensor.” Faria, in turn, has no immediate plans to use his device in health services. “The system needs further studies to achieve a final product that can be used in hospitals and clinics,” he acknowledges. “More research has to be conducted on detecting multiple proteins simultaneously, and the method needs to be automated.”

Emanuel Carrilho, a bioanalytical chemistry expert of the Chemistry Institute of São Carlos, of the University of São Paulo (USP), says the two types of sensors are promising. “Their platforms are different, with different biosensors, but what the two have in common that is very interesting is multiplexing capability, that is, they can detect several biomarkers in a single test,” he says. “Multiplexing will allow a full diagnosis, which will show the presence or absence of cancer and the type of cancer.”

Another important aspect he notes is the use of nanoparticles, which amplify the signals from the devices. For Carrilho, who is also a researcher at the National Institute of Science and Technology of Bioanalytics (INCTBio), the challenge for the new sensors is to have antibodies for all types of cancer.

1. Neoadjuvant treatment for rectal cancer: identification of a genic signature capable of predicting response to treatment and the development of personalized biomarkers for assessing minimal residual disease (nº 2011/50684-8); Grant Mechanism Regular Research Grant; Principal Investigator Anamaria Camargo (LICR); Investment R$361,226.21.
2. Development of biosensors for protein biomarkers to be applied in early detection and monitoring of cancer (nº 2011/02259-6); Grant mechanism Scholarship abroad; Grant Recipient Ronaldo Censi Faria (UFSCar); Investment R$10,780.00 and US$35,400.00.

Scientific articles
CARPINETI, P. et al. The use of personalized biomarkers and liquid biopsies to monitor treatment response and disease recurrence in locally advanced rectal cancer after neoadjuvant chemoradiation. Oncotarget. October 6, 2015.
CAMPESATO, L. F. et al. Comprehensive cancer-gene panels can be used to estimate mutational load and predict clinical benefit to PD-1 blockade in clinical practice. Oncotarget. October 1, 2015.
KADIMISETTY, K. et al. 3D-printed supercapacitor-powered electrochemiluminescent protein immunoarray. Biosensors and Bioelectronics. V. 77, pp. 188-93, March 2016.