Since its emergence in 2019, COVID-19, driven by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), continues to spread at a rapid pace. By March, 2024, the number of deaths reached 7 million worldwide, of which Malaysia recorded 37,349 (https://data.moh.gov.my/dashboard/covid-19). Of note, >1,560 SARS-CoV-2 lineages have been identified [PANGO Lineages network (https://cov-lineages.org/lineage_list.html; updated on December 4, 2024)]. Novel variants of concern (VOCs) with high mutation rates are a challenge. These variants, such as Delta and Omicron, become more infectious and evade vaccines to some degree, hindering pandemic control (1,2).

Of note, five major VOCs have emerged sequentially: Alpha (B.1.1.7/Q.*), Beta (B.1.351/), Gamma (P.1/P.1.*), Delta (B.1.617.2/AY.*) and Omicron (BA.1.1.529/BA.*). Delta, first detected in April, 2021, became dominant, but was then replaced by Omicron in December, 2021. The XBB variant was reported in Malaysia in September, 2022.

The monitoring of new variants through DNA sequencing is crucial for informing public health decisions. Canada leads in sequencing, analyzing almost 13% of positive cases. Despite having the 28th highest number of confirmed cases globally, Malaysia has sequenced <1% of these. (EpiCoV database in GISAID; https://www.gisaid.org/submission-tracker-global/). In developing nations, such as Myanmar, Libya and Azerbaijan, this issue is particularly acute, with sequencing coverage being <0.05%

While DNA sequencing is the gold standard, some researchers propose using reverse transcription-quantitative polymerase chain reaction (RT-qPCR) with specific probes for more rapid and cost-effective variant surveillance (3). However, RT-qPCR may still be costly for some countries. Researchers are exploring affordable point-of-care tests for the diagnosis of COVID-19. RT-loop-mediated isothermal amplification (LAMP) shows promise as a reliable and rapid diagnostic tool (4,5). This method can detect the targeted region in a genome with considerable specificity and sensitivity in <40 min, without the necessity of a real-time thermocycler. The LAMP method utilizes a DNA polymerase and four primers to hybridize six regions on target DNA (6). An inner primer initiates LAMP by recognizing the sense and antisense strands of the target. Strand displacement DNA synthesis, driven by an outer primer, produces a single-stranded DNA template for further synthesis with the second inner and outer primers, forming a stem-loop structure. In the subsequent LAMP cycles, one inner primer binds to the loop of the product, initiating more synthesis and generating both the original and a new stem-loop DNA with an extended stem. Signal can be captured based on the presence of amplification through gel electrophoresis, colour change or fluorescence detection. An improved LAMP method utilizes two extra primers (termed forward loop and backward loop) which targets loop region (7). This six-primer LAMP reaction significantly reduces the signal generation to 15-30 min.

The present study developed RT-LAMP primers aimed at detecting specific mutations found in the Delta, Omicron and XBB variants. This novel assay successfully distinguished between samples containing these variants, providing an indirect method to monitor their prevalence. To the best of our knowledge, this is the first report on the use of RT-LAMP for VOC detection without additional probes or primers. The present study proves that RT-LAMP can be a more cost-effective alternative to DNA sequencing and RT-qPCR for variant surveillance.

Throughout the pandemic, SARS-CoV-2 has been evolving through mutations in its genetic code. These mutations can endow the virus certain advantages, such as spreading more easily or evading our immune defenses (2). Variants with these beneficial mutations, known as VOCs tend to become more prevalent in the population. To remain ahead of the virus, it is crucial to track new variants with the potential to spread rapidly. WGS is a powerful tool for identifying these new variants. However, researchers have also discovered specific mutations that are unique to each VOC. These unique mutations function as a fingerprint, allowing scientists to identify the variant without needing full genetic sequencing.

The geographical distribution of SARS-CoV-2 variants has been uneven, with some regions experiencing rapid and successive waves of different variants, while others have had less data due to limited sequencing. The Delta variant was particularly widespread, predominating in Africa, Asia, Europe, North America and Oceania by July, 2021(9). In South America, the Gamma variant remained dominant for a time, before being overtaken by the Delta variant. The emergence of the Omicron variant then rapidly shifted the global landscape again. However, the detection and tracking of these variants has been heavily influenced by sequencing capacity, with countries such as the United Kingdom having higher sequencing coverage compared to countries such as Brazil, where the sequencing of positive cases is markedly lower (8). This discrepancy in genomic surveillance highlights the challenge of monitoring variants in real-time.

The lack of sequencing capacity in a number of low- and middle-income countries creates a significant gap in the understanding of variant spread and evolution (10). This limitation means that data from these regions may be underrepresented in global databases, such as GISAID, potentially leading to an incomplete picture of global variant dynamics (9). The uneven distribution of sequencing capacity also means that some countries may experience surges in cases due to a variant, but have limited ability to understand which variant is responsible. There is a need for more accessible and affordable methods for variant detection, such as RT-LAMP, to be deployed in these regions to complement limited sequencing capacity and to help ensure effective public health responses and control measures.

While some recent research has focused on using RT-LAMP to detect all SARS-CoV-2 variants regardless of strain (11,12), this approach offers limited value for tracking specific variants in epidemiological studies. However, RT-LAMP has exhibited promise in identifying specific variants. For example, Sherril-Mix et al (13) developed a method using molecular beacons to detect the Alpha variant (B.1.1.7) by targeting the S1Δ69-70 deletion in its genome. Another study by Yang et al (14) presented an RT-LAMP genotyping method that specifically identifies Delta variants by analyzing the R203M mutation in the N gene. This method achieved high accuracy without requiring additional probes, but relied on calculating the Ct ratio between results from two separate reactions per sample. Building on this progress, dos Santos et al (8) proposed a concept for an RT-LAMP assay targeting mutations C21614T (S:L18F) and C21638T (S:P26S) specific to Gamma variants and their descendants. This method avoids complex equipment and additional probes often required in conventional quantitative LAMP protocols. The present study aimed to take this a step further by introducing specific primer sets designed to detect the Delta, Omicron and XBB variants based on their unique mutations (S: L452R, S: T478K, S: E452A and A19326G).

In the present study, a total of 23 sets of LAMP primes were designed and screened for their performance against targeted SARS-CoV-2 variants (i.e. Delta, Omicron and XBB). Only three sets of LAMP primers exhibited promising performance against its target (Table I). This suggests a limited capability of openly available software tools (i.e., PrimerExplorerV5) and a poor understanding in LAMP primer design. One should be prepared to design and screen a handful of LAMP primers to increase the chances of obtaining a few useable candidates. A notable contribution of the present study is that it describes novel RT-LAMP primers that can distinguish between COVID-19 samples carrying mutations specific to the VOCs, Delta, Omicron and XBB (and its descendants). These mutations function as markers for identifying these VOCs with an accuracy of 71 to 92% in samples with a Ct value ≤25). Notably, the assay even performs well for samples with a Ct value as high as 30. This accuracy is comparable to that reported by dos Santos et al (8) (91.11%), highlighting the potential of these primers for SARS-CoV-2 lineage detection.

The recent study by Alhamid et al (15) demonstrated an optimal LAMP performance on SARS-CoV-2 detection by removing outer primer (i.e., LF) in the LAMP reaction. This five-primer LAMP assay avoids misamplification for up to 120 min, thus significantly preventing false positive results. Coincidently, primer set LRTK (Table I) in the present study is also a five-primer LAMP system. Outer primer LF was not suggested by PrimerExplorerV5 for primer set LRTK due to the low scoring of LF candidates. Nevertheless, primer set LRTK was tested and exhibited a good performance against the Delta variant detection. However, LAMP reaction using primer set LRTK was not performed and its misamplification ability was not determined. While the study by Alhamid et al (15) and the present study demonstrated the practicality of five-primer LAMP reaction, further studies need to be performed on different target locations (for the same gene) and different genes to prove its usefulness.

The RT-LAMP assay is proving useful for monitoring emerging disease variants. It offers sensitivity comparable to traditional sequencing, the current gold standard. In the present study, RT-LAMP successfully distinguished the majority of samples with a Ct value ≤25, with some Ct values reaching 30. Sequencing functions best with samples that have lower Ct values as they provide more genetic material, leading to more accurate variant identification. This suggests that the RT-LAMP assay from the present study could be a valuable screening tool in areas with limited sequencing capabilities.

A key advantage of RT-LAMP for variant identification is its speed and affordability compared to other techniques. It has been demonstrated that RT-LAMP costs significantly less per sample, at least 50-fold more economical than next-generation sequencing (NGS) and 4.6-fold more economical than RT-qPCR (8). Furthermore, RT-LAMP boasts a more rapid turnaround time, with results available in 30 min to 1 h. This is a notable improvement over RT-qPCR, which typically takes at least 2 h, and NGS, which can take up to 1 or 2 days. These attributes render RT-LAMP a promising tool for rapid diagnosis and VOC surveillance.

With vaccinations rising globally, keeping track of new variants is crucial. This helps policymakers make informed decisions and prevent outbreaks from highly transmissible or antibody-evading variants, such as Omicron. The present study highlights how a special type of RT-LAMP test, designed to target specific lineages, can be a valuable tool. It allows researchers to estimate how common these lineages are within a population during a specific week. This test could be particularly useful in areas where sequencing resources are limited or expensive. It can act as a preliminary screening method before resorting to more in-depth sequencing techniques.

There are two main ways to improve these assays: Exploring enzymes better suited for displacing existing strands and creating cost-effective, in-house reagents. In fact, a previous study was conducted to evaluate the usefulness of freeze-dried (lyophilized) RT-LAMP reagents as at-home self-testing kit (5). The authors of that study demonstrated that lyophilized RT-LAMP reactions had fewer false positives compared to those using liquid reagents. This technique offers several advantages: An extended shelf life, room-temperature storage and the elimination of cold-chain shipping costs (5).

While RT-LAMP offers significant advantages for SARS-CoV-2 detection, it is essential to acknowledge its limitations. One notable concern is the risk of false positive results, which can arise from non-specific amplification, especially when using colorimetric detection (15). Suboptimal primer design can also lead to misamplification (16). Additionally, sample quality can impact assay performance. In particular, the presence of inhibitors in crude samples such as saliva, blood, and urine can affect the accuracy of the test (12). Some studies have found that RT-LAMP becomes less reliable for samples with high cycle quantification values (Cq) on RT-qPCR. Moreover, visual interpretation of colorimetric results can introduce user errors, and extended storage at elevated temperatures can reduce the colour contrast, increasing misinterpretations (5). Despite these limitations, the ongoing optimization of RT-LAMP assays aims to reduce these issues. For example, the use of five primers instead of six can reduce the false positive rate (15).

Integrating RT-LAMP into current COVID-19 testing strategies, particularly in low-resource settings, requires a strategic approach. Due to its low cost and simplicity, RT-LAMP can be used as a screening tool to identify potentially positive cases. Positive results can then be confirmed using a more accurate test like RT-qPCR, if resources allow. The rapid turn-around time of RT-LAMP makes it valuable for mass surveillance, and this is particularly important for identifying the virus earlier in the transmissible phase. RT-LAMP assays that utilize a simple workflow and minimal equipment are ideal for use in decentralized settings, such as local pharmacies, mobile labs or in the home. In these situations, a technician or the patient themselves would take the sample using a swab and then return the sample for testing. Furthermore, the capacity of RT-LAMP assays to detect variants using specific primers makes it useful for monitoring emerging variants. Overall, the trade-offs between speed, cost and accuracy should be considered when deploying RT-LAMP as a complementary tool to existing methods, particularly in low resource settings.

In conclusion, these novel SARS-CoV-2 variant detection methods are innovative. At under $5 per test (excluding labor and processing), they are significantly more cost-effective than existing options. In addition, they can be easily scaled up or down to fit the needs of any setting, from small workplaces with a handful of employees to schools with a few thousand students. The simple equipment needed and the low cost render them ideal for areas with limited resources, which have often been hit hardest by the pandemic.