Fluorescent and Colorimetric RT-LAMP as a Rapid and Specic Qualitative Method for Chronic Myeloid Leukemia Diagnosis

The detection of BCR-ABL1 mRNA transcripts is essential to molecular chronic myeloid leukemia (CML) diagnosis. In most cases, the RT-qPCR technique is performed as the gold standard diagnosis tool for clinical cases. However, this method requires expensive reagents and equipment, such as a real-time thermal cycler, probes and master mix. Consequently, the development and validation of simple and low-cost methods are essential for a rapid CML diagnosis in less specialized and equipped centers. In this study, we develop and demonstrate an accessible, rapid, and low-cost method using RT-LAMP for BCR-ABL1 detection in both cell lines and CML clinical samples, using colorimetric and uorescent assays. Differently to the Q-LAMP assay described in 2019 by Stella and collaborators, the samples here were analyzed by RT-qPCR and the results were compared to the results obtained by uorescent and colorimetric RT-LAMP. The obtained data indicates that the proposed method here described is a cheaper, robust and specic approach for CML diagnosis with outstanding performance.


Introduction
Chronic myeloid leukemia (CML) is a clonal myeloproliferative disorder characterized by the translocation t(9;22)(q34;q11), that generates BCR-ABL1 fusion gene and protein, the hallmarks of CML 1 . This agerelated neoplasm has an average diagnosis age estimated around 60 years old in western countries, however, studies indicate that in Africa and Asia the diagnostic age is about ten years younger when compared to other countries, in which the higher frequency occurs in men 2,3

. The American Cancer
Society's estimates that in 2020 about 8,450 new cases were diagnosed in the USA with 1,130 deaths and a 5-year survival rate of 70.4% 4 . The Brazilian National Cancer Institute estimated that there were 10.810 new leukemia diagnostics in 2020, in which around 1,620 were CML 5 .
According to the European Society for Medical Oncology, there are standard procedures for CML diagnosis, such as the total baseline assay, differential blood cells count, bone-marrow cytology and karyotype qualitative blood RT-qPCR 6 . CML patients are treated with tyrosine kinase inhibitors (TKIs), one of the most effective oncological targeted therapies. This class of drugs targets the BCR-ABL1 positive cells and induce remission in most cases of CML. However, resistance and tolerance to TKIs can occur, thus, the treatment response needs to be monitored in the determined time-points in order to de ne the appropriate drug therapy, the effective response to the treatment and to select patients who are eligible for TKI discontinuation 7 .
As one of the essential components for CML diagnostics, the identi cation of t(9;22)(q34;q11) BCR-ABL1 transcripts may be performed by qualitative molecular tests, which reduce the complexity and costs at the time of diagnosis, allowing less specialized centers to perform the tests. Real-time PCR is the goldstandard method which indicates the translocation presence, while RT-PCR technique combined with electrophoresis is performed mainly in the scienti c research context 8,9 . BCR-ABL1 transcripts are also quanti ed by real-time PCR to monitor the response to the TKIs treatment, demonstrating its e ciency and indicating which is the most suitable therapy drug 10,11 .
Recently, the Loop-Mediated Isothermal Ampli cation (LAMP) methodology has been studied and applied for several molecular biology diagnosis, demonstrating high sensitivity and speci city aspects [12][13][14][15][16] . This method is based on the use of the strand-displacement Bst polymerase, combined with sets of 4 to 6 primers in an isothermal procedure, usually between 60-65 o C for 30-60 minutes 17,18 . The LAMP methodology is often performed in combination to other detection techniques, such as: electrophoresis in agarose gel with ethidium bromide staining, turbidity based on the precipitation of magnesium pyrophosphate, color changes due to pH alteration and by uorescence through intercalating dyes or probes 19 . This isothermal method can be used to detect both DNA and RNA, plus the Reverse-Transcriptase enzyme can be simultaneously added to the reaction to perform a one-step RT-LAMP. Even though LAMP is manly applied for pathogens detection, it has also a promising application to the human genetic context and disease, such as screening of genetic mutations and genotyping [20][21][22] . In this study we propose a uorescent and colorimetric RT-LAMP for detection of BCR-ABL1 b2a2 and b3a2 isoforms in cell lines and in CML clinical samples, as a fast and speci c approach in clinical practice for CML diagnosis. Using the same RNA dilution templates, three independent experiments were conducted for BCR-ABL1 detection by uorescent RT-LAMP. BCR-ABL1 p210 transcript was detected in the assays at 125, 12,5 and 1,25 ng per well, while the internal control was detected in all concentrations (Fig. 2). When the same experiment was performed using RNA from 10 negative BCR-ABL cell lines, no ampli cation was detected by uorescent RT-LAMP for BCR-ABL1, while internal control was detected in all templates. Table 1 summarizes the results for KCL-22 and K562 RNA templates by RT-qPCR and uorescent RT-LAMP for both BCR-ABL p210 and internal control.  (Fig. 4). Hence, the RT-LAMP described here demonstrated 71.4% of diagnostic sensitivity and 100% of diagnostic speci city.

Results
Interestingly, the comparison between RT-qPCR Ct values and uorescent RT-LAMP results revealed that all BCR-ABL1 p210 positive samples with Ct lower than 28 in RT-qPCR were also positive for uorescent RT-LAMP, demonstrating 100% of sensitivity (Table 2). Lastly, the uorescence of RT-LAMP reactions was quanti ed by spectrophotometry. BCR-ABL1 p210 positive and negative samples showed a signi cant difference in uorescence intensity (p < 0.0001) and were separated into two clear distinct clusters: one with a mean uorescence intensity of 28.527 for positive samples and another with 15.060 for negative samples. There was no difference in uorescence intensity (p = 0.4781) between BCR-ABL p210 positive and negative samples for the IC (Fig. 5).
Colorimetric RT-LAMP: As a rst result, RNA from K562 and A549 were used as template with 125, 12

Discussion
Usually, CML diagnostic is conducted by clinical exams as white blood cells count and classical karyotyping. This last-mentioned method detects large genetic translocation in leukocyte DNA but represents a laborious procedure with operator-dependent e ciency, hence, it is more di cult to be performed in undeveloped countries. Molecular diagnostics are considered gold standard methods for their e ciency, cost and practical procedures when compared to conventional methodologies. The molecular detection of BCR-ABL1 in CML patients not only con rms the diagnostic, but it also represents a biomarker for disease progression as well as a treatment predictor with TKIs.
Several protocols for BCR-ABL detection by RT-qPCR are available in scienti c literature and many international researches have performed this assay standardization [23][24][25][26][27][28][29] . The sensitivity of RT-qPCR for BCR-ABL quanti cation assays is increasing in order to detect lower numbers of leukemia cells and minimal residual disease (MRD). However, these molecular approaches are currently expensive and inaccessible in large scale, especially for low-income countries and patients. Still, leukemia cell counts are usually high in nontreated CML patients, which indicates that these extremely sensitive and expensive tests are not necessary for diagnosis.
In the overall cancer context, screening of EGFR gene mutations at codons 746-753 by LAMP has already been reported, in which the ampli cation of mutated DNA fragments were effective within 30 minutes 30 . 31 developed a LAMP method to identify the EGFR mutation with a comparable e ciency to the Therascreen quantitative PCR assay, suggesting that it is a valuable alternative for the identi cation of oncogene mutations. In a rst report by 32 , the authors designed an allele-speci c LAMP for EGFR T790M mutation optimized with six sets of reaction mixture. According to the results, only one condition could accurately distinguish between the wild-type and the mutated DNA. A second report from the same research group demonstrated a cost effective and sensitive LAMP for detecting therapeutically-relevant EGFR hotspot mutations in a non-invasive manner. The results consistency when compared to those obtained by ultrasensitive droplet digital PCR (ddPCR) assay indicates the robustness of the developed method 33 .
Some interesting reports emerged demonstrating the application of LAMP in the screening of mutant KRAS in oncological samples. 34  RT-PCR following the BIOMED-1 protocol 37 . Differently from the data here described, Q-LAMP uses RT-PCR analyzed in agarose gel as gold standard, not RT-qPCR, that is a more accurate approach. This difference is important because the rst method relies just in an end-point analysis with qualitative performance and does not determine the relative BCR-ABL translocation quantity by returning a Ct value. The BCR-ABL detection by RT-LAMP described in this study gives more details and depth regarding assay performance than the Q-LAMP reported method. Another signi cant difference was that the authors enrolled newly diagnosed patients, usually with a high leukemia cells count, while we accessed patients with low leukemia cells amount.
The colorimetric and uorescent RT-LAMP described here represent qualitative assays for detecting BCR-ABL1 most found isoforms, that performs from the reverse transcription of RNA to the isothermal ampli cation of BCR-ABL1 in a single tube. The internal control is ampli ed in a separated reaction tube. Colorimetric and uorescent RT-LAMP showed 96.43% and 100% of speci city, respectively, and concordance with RT-qPCR in paired BCR-ABL1 negative clinical samples. The assay is adequate for screening diagnosis of BCR-ABL1 but it does not discriminate BCR-ABL1 isoforms and it is not indicated for molecular monitoring by BCR-ABL1 quanti cation. This novel test has advantages over RT-qPCR for CML diagnosis because it decreases the time to obtain the results and most importantly, it reduces the costs and complexity. In conclusion, this two method variation showed a reliable and e cient approach for BCR-ABL1 detection in CML patients and, in uorescent approach can be performed not only on realtime PCR systems but also in conventional thermal cycler followed by a uorescence spectrophotometer, allowing a broader spectrum of possibilities for its use in smaller and less specialized centers. Colorimetric RT-LAMP is even more practical due to no need of complex equipment, just a conventional PCR system, and naked-eye result visualization, with close to 100% of diagnostic speci city and high diagnostic sensibility Methods Samples: RNA from two BCR-ABL1 p210 positive cell lines, K562 and KCL-22 (ATCC -USA) were used for the conceptual validation, analytical sensitivity and for the detection limit of uorescent RT-LAMP. To assess analytical speci city, RNA from 10 BCR-ABL1 p210 negative cell lines were used: Jurkat, HL60, ReH, THP-1, Panc1, AsPC1, SW20, MiaPaCa, Capan and A549 (ATCC -USA).
The sensitivity and speci city for uorescent and colorimetric RT-LAMP approaches were evaluated with 39 and 64 CML clinical samples from patients in chronic phase during TKI treatment, respectively. These samples were collected in Erasto Gaertner Hospital, Curitiba, Paraná, after Erasto Gaertner Hospital Ethics Committee approval (CAAE 08809419.0.0000.0098). All samples collection and experimental conduction were carried out in accordance with guidelines and Brazilian regulations. The project was described in detail for all participants, which have read, discussed, and signed an informed consent before sample collection. For each patient, 4 mL of peripheral blood was collected in EDTA tubes and processed in less than 24 hours. The blood was centrifuged (500g/10 minutes/4°C), and the buffy coat was collected by a pasteur pipet in a 15 mL tube. The RNA extraction was processed utilizing a Qiamp RNA Blood Mini Kit (Qiagen) following the manual instructions. Personal and clinical data such as age, gender, diagnostic date, and treatment were also assessed from patients' charts.
Real-time PCR: To detect BCR-ABL1 p210 in cell lines and clinical samples, RT-qPCR was performed using speci c primers and a probe labeled with HEX for b3a2 and b2a2 BCR-ABL1 p210 isoform. For the internal control (IC), speci c primers and a probe were labeled with FAM. Reactions were conducted using GoTaq® Probe