Background

Malaria remains a leading public health concern in Cameroon, accounting for 29.1% of health facility consultations and 17.2% of deaths in 2020 [1, 2]. Children under five years of age and pregnant women are disproportionately vulnerable. In 2020, hospital morbidity due to malaria was 40.1% among children under five years and 22.5% for pregnant women [1, 2]. In the last two decades, efforts and progress have been made worldwide to control the disease by implementing several vector control measures in addition to therapeutic care. The use of insecticide-treated nets (ITNs) has contributed to the drastic reduction of the disease burden [3, 4]. Nonetheless, sub-Saharan Africa is still at risk and has the most malaria cases and deaths worldwide. According to the 2021 World Malaria Report, there were an estimated 241 million malaria cases recorded in 2020 globally—an increase from 227 million cases in 2019—with the majority of the increased cases reported from countries in the WHO African Region [3, 4].

In Cameroon, the National Malaria Control Programme (NMCP) and its partners have implemented a three-pronged malaria response, including: (i) free distribution of ITNs through mass campaigns and during antenatal consultations for pregnant women, (ii) seasonal malaria chemoprevention for children aged 3 to 59 months, specifically in the North and Far North regions, and (iii) free treatment of uncomplicated and severe malaria for children under five and subsidized case management of malaria for the general population by supporting any malaria case diagnosis and treatment. The country started implementing ITN mass distribution in 2011 and has conducted three mass campaigns (2011, 2015, and 2019) using pyrethroid-only ITNs in all regions. However, the effectiveness of these control measures is being threatened by factors such as the resistance of vectors to the insecticides used in different ITNs [5, 6], the change in vector behaviours, human population behaviour and movement throughout the night [7, 8, 9, 10], and/or the resistance of the Plasmodium falciparum parasite to anti-malarial drugs [11, 12]. As in most sub-Saharan African countries, pyrethroid resistance involving different target sites and metabolic resistance within the main vector Anopheles gambiae sensu lato (s.l.) population is widespread in Cameroon [13, 14, 15, 16]. Furthermore, several potential malaria vectors have emerged as a new challenge for vector control because the existing interventions target specific indoor feeding and resting behaviours of the primary vector species. Cameroon is at particular risk given it hosts several species of Anopheles mosquitoes that have been found to carry malaria parasites [12].

The complex vector-parasite ecology in Cameroon requires that malaria control efforts consider all vector species rather than target a single malaria vector. Continuous evaluation of vector bionomics in a changing landscape is required to improve the vector control strategy. While several entomological studies have been conducted in the country to describe malaria transmission parameters, these are often conducted within a short time frame or in a limited number of sites [12, 17,18,19]. Based on this context, and to provide recent and extensive entomological data to the NMCP, the U.S. President’s Malaria Initiative (PMI) VectorLink project conducted vector surveillance from 2018 to 2020 in five sentinel sites representing four ecological zones in five of the 10 regions of the country. Vector bionomics and malaria transmission parameters were assessed to support the country’s vector control strategy, including the selection and deployment of appropriate evidence-based vector control tools.

Methods

Study sites

Cameroon is subdivided into 10 health regions (Southern Region, East, Centre, Littoral, South-West, West, North-West, Adamaoua, North, and Far North Region). Malaria endemicity in the country varies by region, with the highest incidence in the East Region with 180 confirmed cases per thousand inhabitants in 2021 to the lowest in the Far North Region (< 50 cases per 1000). However, the highest morbidity was recorded in the North with more than 40 deaths due to malaria in 2021; the incidence is also high in this region [20]. Two sites (Gounougou and Simatou) were selected in the northern part of the country and three (Mangoum, Nyabessang, and Bonabéri) in the southern part. Gounougou (13.55°E; 9.07°N) is a rice cultivation area located in the dry savannah zone of the North. It has a rainy season of about six months (May to October) and is in one of the high malaria endemicity regions and where the highest morbidity occurred. Simatou (15°E; 10.34°N), situated in the Sahelian zone in the Far North, is also a rice cultivation area, with a short rainy season occurring from July to October and a moderate malaria incidence (119.0 cases per 1000).

Mangoum (10.58°E; 5.47°N) is in the wet savanna zone in the West region of the country where malaria endemicity is relatively low (96.0 cases per 1000). Nyabessang (10.39°E; 2.4°N), a rural area located in the forest of the South region with high rainfall, is surrounded by many rivers and dams and has low malaria endemicity (94.3 cases per 1000). The final site, Bonabéri (9.65°E; 4.08°N) is an urban area located in the coastal zone of the Littoral region with a low endemicity (92.2 cases per 1000) (Fig. 1).

Fig. 1
figure 1

Map of Cameroon showing the geo-location of the five vector surveillance sentinel sites and vegetation across the country

Full size image

Vector bionomics monitoring

The study was conducted in the five sentinel sites from October 2018 to September 2020. Monthly entomological data collections were carried out in Gounougou and Simatou from October 2018 through September 2020 (except for November 2019 through March 2020 when collections were conducted every other month at both sites). Thus, a total of 19 collection-months were completed in these two sites over the survey period. In the southern sites, collections were done every other month in Mangoum and Nyabessang from October 2018 through February 2020 and from December 2018 through February 2020 in Bonabéri. Thereafter, the collections were conducted monthly from June to September 2020 at these three sites. A total of 13 months of field collections were completed in Mangoum and Nyabessang and 12 months in Bonabéri. No collections were conducted between April and May 2020 due to the COVID-19 pandemic. Collections were adjusted to every other month per activity programming including expansion of insecticide resistance monitoring sites across the country. For all collection methods, the same houses were used monthly and throughout the collection period.

Adult mosquitoes were collected in each of the sites using three collection methods: human landing catches (HLCs) which target human host seeking biting vectors, U.S. Centers for Disease Control and Prevention (CDC) light traps which target a diversity of host-seeking species, and pyrethrum spray catches (PSCs) which target indoor resting vectors. Each method enabled the estimation of different parameters which characterized the vector behaviour and malaria transmission within the local populations.

HLCs were conducted in three randomly selected houses per site that were maintained throughout the collection period. Adult mosquito collections were done from 6:00 p.m. to 8:00 a.m. for two consecutive nights per collection month. Four collectors were assigned to each house (two collectors indoors and two collectors outdoors). The collectors were monitored for malaria symptoms before and after collection, and any malaria cases were given treatment. For efficiency and accuracy, two teams of 12 collectors each worked in shifts (from 6:00 p.m. to midnight and from midnight to 8:00 a.m.) and rotated each day. Additionally, the collectors rotated positions every hour throughout the night to account for variation in attractiveness among collectors. The collectors used hemolysis tubes to catch mosquitoes landing on their lower exposed limbs. Mosquitoes were collected hourly and put in separate bags. After each night of collection, mosquitoes were identified morphologically using taxonomic identification keys [21,22,23]. For each collection month and site, a subsample of randomly selected vectors underwent ovary dissection for parity rate determination [24].

CDC light traps were set indoors and outdoors from 6:00 p.m. to 6:00 a.m. in four houses for two consecutive nights per collection month. The indoor traps were baited and suspended nearby a bed with a mosquito net where household members slept. Outdoor traps were hung on a tree with no bait. All traps were suspended 1.5 m above the ground.

PSCs were carried out during two consecutive days per month in 20 houses (10 houses per day) between 6:00 a.m. and 8:00 a.m. A room in which inhabitants spent the night was selected in each house. A white sheet was placed in the room covering the floor and bed. A pyrethroid insecticide spray containing piperonyl butoxide (PBO) synergist was used to spray the room and to collect all indoor resting mosquitoes. When the house had open eaves, these were sprayed first from outside before spraying indoors to prevent the mosquitoes from escaping. After about 10 min post-spraying, the sheets were gently brought outdoors and the mosquitoes on the sheets were collected using forceps and preserved in petri dishes for morphological identification. The abdominal status of the collected vectors was determined, and the number and percentage of blood-fed mosquitoes were recorded during morphological identification [21].

Parity assessment

To determine the parity rate of Anopheles species collected, approximately 20% of unfed, female Anopheles collected using HLCs were randomly selected each month for ovary dissection, following the methods described by Detinova 1962, by observing the degree of coiling by the ovarian tracheoles [24]. All Anopheles and the carcasses of the dissected Anopheles were individually stored in labeled Eppendorf tubes containing silica gel for further molecular analysis.

Molecular characterization

A random subsample of about 100 field preserved mosquitoes that were morphologically identified as either An. gambiae s.l. and Anopheles funestus s.l. (when collected) were selected per month, per collection method, and per site, and used for molecular species identification using polymerase chain reaction (PCR) methods to discriminate between sibling Anopheles species using wings and legs. Additionally, about 400 mosquitoes were randomly selected from each of the Anopheles species collected to detect sporozoite infections using the head and thorax using indirect enzyme-linked immunosorbent assay (ELISA), and 100 abdomen of PSC-collected An. gambiae s.l. were sampled per month to determine blood meal sources using direct ELISA.

Genomic DNA extraction and species identification

Whole genomic DNA (gDNA) was extracted from each mosquito sample following the LIVAK method [25] and stored at − 20 ºC. A NanoDrop™ spectrophotometer (Thermo-Scientific, Wilmington, USA) was used to determine the concentration and purity of the extracted DNA.

Members of An. gambiae complex (An. gambiae sensu stricto (s.s.), Anopheles coluzzii, and Anopheles arabiensis) were discriminated using the Short Interspersed Nuclear Element (SINE) PCR protocol of Santolamazza et al. [26]. In the coastal sites such as Bonabéri and Nyabessang, where other species, such as Anopheles melas are present, the PCR-RFLP protocol described by Fanello et al. [27] was used to discriminate the An. gambiae species complex. Anopheles funestus group subspecies were characterized using a cocktail PCR according to Koekemoer et al. [28] with the addition of the Anopheles rivulorum-like primers. Genomic DNA from 100 randomly selected mosquitoes were processed per month and per site. PCR products were run via electrophoresis through a 1.5% agarose gel with Midori Green® (Gene flow, UK) and visualized under ultraviolet light.

Circumsporozoite infection detection

Determination of sporozoite infection rates and blood meal analysis of adult Anopheles mosquitoes collected using HLCs and PSCs were conducted using circumsporozoite ELISA (csELISA) following the method described by Burkot et al. [29] and modified by Wirtz et al. [30] for sporozoite detection in the head and thorax of mosquitoes. This method uses a monoclonal antibody that recognizes a repetitive epitope on the circumsporozoite protein of P. falciparum. Plasmodium falciparum sporozoite ELISA reagent kits (MRA-890) were obtained from BEI Resources (NIAID, NIH, USA). Lyophilized P. falciparum monoclonal antibody was reconstituted prior to utilization using a glycerol-water solution to achieve a final concentration of 0.5 mg/ml. Similarly, all reagents including phenol red, 1X Phosphate Buffered Saline (PBS), Blocking Buffer (BB), Grinding Buffer, and 1X PBS-Tween wash solution were prepared before starting the manipulation and according to the manufacturer guidance (MR4-890 kit). Diluted P. falciparum sporozoite recombinant proteins supplied by CDC (Atlanta, USA) were used as positive controls, while ground male mosquitoes were used as negative controls. Determination of positive samples was done after reading optical densities (OD) at 405 nm on an ELISA plate reader (Biotek Elx800, Swindon, UK). Positive samples were determined by OD readings two-fold greater than the negative controls and were tested a second time for validation.

Identification of blood meal source

The source of the blood meal contained in the abdomen of resting mosquitoes collected by PSCs was determined using a direct ELISA technique described by Beier et al. [31]. This technique allows the identification of human, cow, pig, chicken, goat, horse, and dog blood. Peroxidase conjugated antibodies and animal heterologous serum were obtained from Sigma (St. Louis, USA). After manipulation, absorbance at 414 nm was determined using an ELISA plate reader. Samples were considered positive if the absorbance values exceeded the mean plus three times the standard deviation of four negative controls represented by unfed mosquitoes.

Vectorial capacity of Anopheles species

The vectorial capacity represents the ability of a population of vectors to transmit Plasmodium spp. in terms of the potential number of secondary inoculations originating per day from an infective person. The vectorial capacity is dependent upon a series of biological characteristics such as population density, blood meal preference, and the probability of the vector to survive per day. The MacDonald formula was used to estimate the vectorial capacity of each Anopheles species found with P. falciparum parasite across all sentinel sites assuming that an infectious person will be subject to m mosquito bites (assuming everyone is equally attractive) and will receive ma bites each day corresponding to the HBR. For those mosquitoes to become infectious they must survive the extrinsic incubation period (with probability pn). The adult mosquitoes (on average) live for 1/(− ln (p)) days biting, and potentially infecting, humans at a rate of “a” per day [32]. The equation combines these quantities to give the total potential infectious bite index arising from one infected person for one day, as described below:

\(VC=\frac{{\left(m{a}^{2}\right)p}^{n}}{- ln\left(p\right)}\)Where a = the ratio of mosquitoes feeding on human, m = ma (man aggressivity) represented by the HBR of the vector, the parasite’s extrinsic incubation period (EIP, n days) which we considered as 12 days, and p = the mosquito’s survival through one day calculated using the parity rate.

Data management and statistical analysis

All entomological data was regularly entered in Epi-Info Version 3.5.4 by a database manager to facilitate analysis. The proportion of each identified mosquito species was calculated as a percentage of each species out of the total Anopheles collected. The mean IRD of each Anopheles species collected using PSCs was calculated monthly by dividing the total number of mosquitoes collected by the total number of houses visited. The sporozoite infection rate, measured as the proportion of mosquitoes found with circumsporozoite antigen by ELISA, was calculated by dividing the number of positive mosquitoes by the total number of mosquitoes tested per month. The mean sporozoite rate represents the average of all monthly infection rates per site and per species tested. The mean parity rate was determined by dividing the number of parous females by the total number dissected and averaged over the collection period for the total mean parity rate per site.

The EIR was calculated as the product of the HBR and circumsporozoite antigen rate as determined by ELISA per month, while the mean EIR over the study period represents the average of all monthly EIRs per site and per species. The indoor and outdoor EIRs were compared using the Mann-Whitney test of XLSTAT software for comparing the mean of two series of numbers at 5% significance level. The human blood index (HBI) was calculated as the proportion of mosquitoes found to contain human IgG by ELISA out of the total mosquitoes tested.

Results

Anopheles mosquito species composition

Overall, 139,326 Anopheles mosquitoes representing 18 distinct species were collected in the five sentinel sites using the three methods (HLC, CDC light trap, and PSC). A total of 83,540 Anopheles mosquitoes (60.0%) were collected using HLCs; 29,846 (21.4%) were collected with CDC light trap; and 25,940 (18.6%) were collected by PSC. Anopheles gambiae s.l. (98,867; 71.0%) was the predominant species and was collected in all sites using the three methods. Anopheles moucheti and Anopheles nili were only found in Nyabessang (Table 1 and Additional file 1: data 1).

Table 1 Species composition of Anopheles mosquitoes collected* by all methods across five sites in Cameroon
Full size table

A total of 5598 An. gambiae s.l. mosquitoes from all five sites were DNA extracted for species identification. Of these, 283 (5.1%) did not amplify, while 5315 An. gambiae s.l. (1513 from Simatou, 1,538 from Gounougou, 962 from Mangoum, 575 from Nyabessang, and 727 from Bonabéri) and 596 An. funestus s.l. (217 from Simatou, 368 from Gounougou, 9 from Mangoum, and 2 from Nyabessang) were successfully tested by PCR for molecular identification of the sub-species of each complex (Table 2 and Additional file 1: data 1). Three species from the An. gambiae complex were identified in Simatou and Gounougou including An. gambiae s.s. (1.1% in Simatou, 2.4% in Gounougou), An. coluzzii (90.0% in Simatou, 84.2% in Gounougou), and An. arabiensis (8.7% in Simatou and 13.4% in Gounougou). Hybrids of An. gambiae/An. coluzzii (0.2%) were also recorded in Simatou.

Table 2 Species composition of Anopheles gambiae complex and An. funestus group collected across five sites in Cameroon
Full size table

In the three southern sites, two species of the An. gambiae complex were recorded, including An. gambiae s.s. (98.9% in Mangoum, 93.5% in Nyabessang, and 1.4% in Bonabéri) and An. coluzzii (0.6% in Mangoum, 5.2% in Nyabessang, and 98.6% in Bonabéri). A small proportion of hybrids of both species was also recorded in Mangoum (0.4%) and Nyabessang (1.2%).

For An. funestus s.l., two species of the group were found in Simatou and Gounougou: An. funestus s.s. (59.4% and 89.7%, respectively) and Anopheles leesoni (40.5% and 10.3%, respectively). Prior to molecular identification, a second morphological identification was conducted by the laboratory team to ensure that the tested samples were An. funestus s.l. to avoid misidentification of the An. leesoni that can occur [33] and all detected An. leesoni were re-run using a mix with An. gambiae primers to confirm the species.

Malaria transmission parameter’s estimates

Malaria vectors and human biting rates

The HLC method was the most productive collection method at all sites. Of the 83,540 Anopheles mosquitoes collected, An. gambiae s.l. represented the main vector species in all sites, except in Nyabessang, where An. moucheti and Anopheles paludis were predominant. The mean HBR of Anopheles mosquitoes collected using HLC varied across the sites: 101.3 bites/human/night (b/h/n) in Simatou, 39.2 b/h/n in Gounougou, 15.9 b/h/n in Mangoum, 30.5 b/h/n in Nyabessang, and 17.5 b/h/n in Bonabéri. The HBR of An. gambiae s.l. was 73.9 b/h/n in Simatou, 35.6 b/h/n in Gounougou, 15.6 b/h/n in Mangoum, 3.5 b/h/n in Nyabessang, and 17.5 b/h/n in Bonabéri (Fig. 2 and Additional file 2; data 2). Anopheles gambiae s.l. exhibited a similar biting pattern both indoors and outdoors throughout the night in all sites with the highest biting recorded between 11:00 p.m. and 5:00 a.m. Simatou and Gounougou recorded the highest mean hourly peak biting with 8.1 bites per human per hour (b/h/h) and 4.5 b/h/h, respectively.

Fig. 2
figure 2

Mean monthly human biting rates of Gounougou and Simatou (Northern sites) over the collection period (October 2018-September 2020)

Full size image

Trends in the monthly mean HBR of An. gambiae s.l. over time differed in the southern and the northern sites. In Gounougou, the lowest mean HBRs were recorded between October 2018 and January 2019 and in May and June 2019. Two peaks were observed in February (69.4 b/h/n) and August 2019 (106.0 b/h/n). In Simatou, the lowest mean HBRs were recorded from October 2018 to February 2019 and in May 2019, while the peak mean HBRs were observed in July 2019 (281.2 b/h/n) and July 2020 (327.2 b/h/n) (Fig. 2). The mean monthly HBR of An. gambiae s.l. was 15.6 b/h/n in Mangoum with the lowest mean HBR recorded in August 2020 (3.2 b/h/n) and the highest peak in April 2019 (42.9 b/h/n). In Nyabessang, the highest and only peak of An. gambiae s.l. was 14.0 b/h/n and was recorded in December 2018. From February to August 2019, the HBR of An. gambiae s.l. recorded was much lower (between 2.0 b/h/n and 4.8 b/h/n).

Furthermore, peak biting of the predominant An. paludis (34.5 b/h/n) and An. moucheti (34.3 b/h/n) in Nyabessang was recorded in February 2019. During the second year, a peak HBR of 19.3 b/h/n was recorded for An. moucheti in June 2020 and 10.3 b/h/n for An. moucheti in August 2020, showing a replacement of vector population with An. gambiae s.l. which recorded its lowest density during the same period. In Bonabéri, An. gambiae s.l. biting peaked three times in the study period—first in February 2019 (32.7 b/h/n), then in August 2019 (39.0 b/h/n), and finally in July 2020 (35.0 b/h/n). The lowest HBR was observed in December 2019 with 1.7 b/h/n (Fig. 3 and Additional file 2: data 2).

Fig. 3
figure 3

Mean monthly human biting rates of Mangoum, Nyabessang and Bonabéri (Southern sites) over the collection period (October 2018–September 2020)

Full size image

In addition to the main malaria vector An. gambiae s.l., the mean HBRs of all other potential vectors were also estimated per site. In the three sites where An. funestus s.l. were collected, an average HBR of 0.4 b/h/n was recorded in Simatou, 2.2 b/h/n in Gounougou, and 0.1 b/h/n in Mangoum. The mean HBR of Anopheles ziemanni was 2.3 b/h/n in Simatou, 0.3 b/h/n in Gounougou, and 0.4 b/h/n in Mangoum. Anopheles paludis and An. moucheti were mostly found in Nyabessang with mean HBRs of 13.2 b/h/n and 10.5 b/h/n, respectively. Anopheles nili and Anopheles marshallii had mean HBRs of 2.0 b/h/n and 0.6 b/h/n, respectively. Anopheles pharoensis and Anopheles demeilloni were only collected in the northern sites, with 18.6 b/h/n and 0.5 b/h/n for An. pharoensis in Simatou and Gounougou, and An. demeilloni with a biting rate of 6.0 b/h/n in Simatou. Other potential vectors collected included Anopheles coustani and Anopheles welcomei in Simatou, and Anopheles tenebrosus, Anopheles smithii, and Anopheles christyi in Gounougou. (Table 3 & Additional file 1: data 1).

Table 3 Total mean human biting rate of Anopheles mosquitoes collected using human landing catches from Oct 2018 to Sept 2020 across five sites in Cameroon
Full size table

Outdoor mean HBRs were slightly higher than indoors, but not significantly different in any sites. The mean indoor and outdoor HBRs of An. gambiae s.l. were 34.3 b/h/n and 36.9 b/h/n, respectively in Gounougou, 75.8 b/h/n indoors and 72.0 b/h/n outdoors in Simatou, 16.0 b/h/n indoors and 15.2 b/h/n outdoors in Mangoum, 3.4 b/h/n indoors and 3.7 b/h/n outdoors in Nyabessang, and 10.8 b/h/n indoors and 24.2 b/h/n outdoors in Bonabéri. The endophagic rate of An. gambiae s.l. was 50% in Simatou, 47% in Gounougou, 53% in Mangoum, and 49% in Nyabessang. Only An. gambiae s.l. from Bonabéri were found to bite more outdoors than indoors with an endophagic rate of 30%. The same trends were observed with An. funestus s.l. in Simatou (53%) and Gounougou (57%). A mean endophagic rate of about 50% was also recorded for all other Anopheles collected in specific sites.

Entomological inoculation rate

A total of 15,944 Anopheles mosquitoes including 11,199 An. gambiae s.l. and 4745 other Anopheles were tested by ELISA, of which 496 (402 An. gambiae s.l. and 94 other Anopheles) were found with the Plasmodium circumsporozoite antigen, for a total average infection rate of 3.1%. Twelve Anopheles species were found with Plasmodium parasites including An. gambiae s.l., An. funestus s.l., An. nili, An. moucheti, An. demeilloni, An. pharoensis, An. ziemanni, An. multicinctus, An. marshallii, An. tenebrosus, An. rufipes, and An. paludis. The infection rates recorded across sites were as follows: Gounougou (17.0%), Simatou (13.0%), Mangoum (11.0%), Nyabessang (27.0%), and Bonabéri (4.0%). Anopheles gambiae s.l. from all five sites, as well as An. ziemanni from the four sites where it was collected, tested positive for the Plasmodium circumsporozoite antigen (Table 4).

Table 4 Entomological inoculation rate of Anopheles mosquitoes collected across five sites in Cameroon from October 2018 to September 2020
Full size table

The total mean entomological inoculation rates (EIRs) varied from 18.1 infected bites/human/month (ib/h/m) in Bonabéri to 99.0 ib/h/m in Simatou. Gounougou and Mangoum recorded the second highest EIRs with 55.4 ib/h/m and 51.2 ib/h/m, respectively. Among all species, An. gambiae s.l. contributed the most to malaria transmission in all sites. Furthermore, the total mean EIRs of An. gambiae s.l. were slightly higher outdoors than indoors in Simatou (94.1 ib/h/m vs. 80.9 ib/h/m, p = 0.381), Mangoum (58.1 ib/h/m vs. 46.3 ib/h/m, p = 0.771), and Bonabéri (25.5 ib/h/m vs. 10.6 ib/h/m, p = 0.118), but were not significantly different at any site. Gounougou and Nyabessang recorded similar mean indoor and outdoor EIRs of 53.8 ib/h/m and 51.0 ib/h/m (p = 0.768), and 6.7 ib/h/m and 5.9 ib/h/m, p = 0.408), respectively (Additional file 2: data 2). At least two Anopheles species were involved in malaria transmission in four of the sites (Table 4). Simatou and Nyabessang recorded the largest number of malaria vectors with six Anopheles species involved in the transmission of the parasite. Gounougou recorded four vectors, while in Bonabéri, the only malaria vector found was An. gambiae s.l. The monthly indoor and outdoor HBRs recorded throughout the collection period, coupled with the EIRs, showed that the higher transmission period did not always coincide with the higher biting period in the southern part of the country. EIRs were highest when densities were relatively low in June 2019 in Mangoum and Bonabéri and in August 2019 in Nyabessang and all three southern sites. In contrast, EIRs in the northern sites of Simatou and Gounougou peaked when densities were highest in July and August 2019, respectively.

Parity rate

Across all five sites, ovaries of 11,051 Anopheles samples were dissected during the collection period. The average parity rate across all sites was 68.9%, ranging from 57.1% (Nyabessang) to 76.4% (Gounougou) (Table 5). The mean parity rate of An. gambiae s.l. in Gounougou (75.1%) was significantly higher than that of the four other sites (chi2 = 201.3, ddl = 3, p < 0.00001). However, all Anopheles species dissected showed high parity rates across all sites (Additional file 3: data 3).

Table 5 Parity rate of the Anopheles mosquitoes across five sites in Cameroon
Full size table

Malaria vector resting behaviour

Eleven Anopheles species were collected resting indoors using PSCs representing 18.6% (25,940) of the total Anopheles mosquitoes collected at all sites during the collection period. Similar to HLCs, Simatou and Gounougou recorded a more diverse and higher number of different Anopheles species collected through PSCs compared to other sites. Seven of the 11 species collected overall were found in the two northern sites and included An. gambiae s.l., An. funestus s.l., An. ziemanni, An. rufipes, An. pharoensis, An. hancocki, and An. demeilloni. Two An. multicinctus were also collected in Gounougou while An. moucheti and An. nili were found in Nyabessang. In Mangoum and Bonabéri, An. gambiae s.l. was the only species collected.

Indoor resting density across sites

The average density per room of Anopheles mosquitoes resting indoors (IRD) was 17.1 females/room (f/r) (25,940 total females/1520 rooms visited). Table 6 describes the IRD per site. The highest mean IRD was recorded in Simatou (39.6 f/r) in the north, while the lowest was observed in Bonabéri (0.04 f/r) in the south and varied by month and season. The mean IRD of An. gambiae s.l. was 34.7 f/r in Simatou and 23.4 f/r in Gounougou. The highest was observed in July 2020 (135.2 f/r) in Simatou and in July 2019 (77.5 f/r) in Gounougou. In the southern sites, the mean IRDs were low compared to the northern sites with a mean of 1.8 f/r in Mangoum, 0.6 f/r in Nyabessang, and 0.04 f/r in Bonabéri. Seasonal variation was also observed in the southern sites where the highest IRD was recorded in October 2018 (4.3 f/r) in Mangoum, in July 2020 (0.2 f/r) in Bonabéri, and in September 2020 (1.2 f/r) in Nyabessang (Additional file 3: data 3).

Table 6 Mean indoor resting density of Anopheles mosquitoes collected by pyrethrum spray catches across five sites from October 2018 to September 2020
Full size table

Host preference of Anopheles species across sites

Nine of the 18 Anopheles species collected from the five sites were screened for blood meal sources to detect if the bloodmeal taken was from a human, cow, sheep, chicken, pig, or horse. A total of 2994 blood-fed Anopheles mosquitoes were analyzed using ELISA, including 2144 An. gambiae s.l., 225 An. funestus s.l., 252 An. rufipes, 83 An. demeilloni, 31 An. pharoensis, 246 An. ziemanni, 2 An. moucheti, 1 An. nili, and 10 An. hancocki. Only 1,151 of the blood-fed mosquitoes analyzed were found to have fed on humans, giving a HBI of 38.4%. The overall HBI varied from 34.3% in Gounougou to 82.3% in Mangoum. The HBIs of An. gambiae s.l. in Mangoum (82.3%) and Nyabessang (64.2%) located in the south were significantly higher than those in the two northern sites (chi2 = 14.18, ddl = 3, p < 0.000001). For An. funestus s.l., the HBI was 44.4% among the samples collected in Simatou and Gounougou. Out of the 252 An. rufipes tested in Simatou and Gounougou, only 12 (4.8%) were found with human blood meal while An. pharoensis tested in Simatou showed a HBI of 40% (12/30).

Vectorial capacity

The vectorial capacity, described as the ability to serve as a vector, was determined for seven species (An. gambiae s.l., An. funestus s.l., An. ziemanni, An. rufipes, An. pharoensis, An. demeilloni, and An. welcomei) in Simatou, six in Gounougou (An. gambiae s.l., An. funestus s.l., An. ziemanni, An. rufipes, An. pharoensis, and An. multicinctus), five in Nyabessang (An. gambiae s.l., An. ziemanni, An. marshallii, An. moucheti, and An. nili) and only An. gambiae s.l. in Mangoum and Bonabéri. Anopheles gambiae s.l. showed the highest vectorial capacity in all sites except Nyabessang, where An. moucheti represented the main potential malaria vector with an index of vectorial capacity of 2.49 versus 0.54). Vectorial capacity in Simatou (32.0) and Gounougou (19.02) indicate a higher capacity of An. gambiae s.l. to transmit malaria compared to the other Anopheles species reported with sporozoite infections. In contrast, An. gambiae s.l. was the main vector collected in Mangoum (9.81) and Bonabéri (10.12) with similar index on vectorial capacity (Table 7). Anopheles funestus s.l. was the second highest contributor of persistent malaria in Gounougou, while An. pharoensis and An. demeilloni represented the two secondary vectors in Simatou (Additional file 4: data 4).

Table 7 Vectorial capacity of Anopheles species collected across five sites in Cameroon
Full size table

Discussion

Entomological vector surveillance is key to describing vector populations and behaviour and thereby informing the development of appropriate vector control strategies and tailored deployment of tools. This study, conducted in different ecological and geographical areas of Cameroon, indicated a high diversity and density of Anopheles species in the country. The longitudinal vector monitoring conducted over about two consecutive years provides concrete density, diversity, and transmission trends of the different malaria vectors across the country. Of the 21 Anopheles species and sub-species collected, 12 species were found to be positive for P. falciparum sporozoites. Other recent studies conducted in Cameroon have revealed a high diversity of malaria vectors distributed across different geographical locations within the country [12, 17,18,19]. However, An. gambiae s.l. was the dominant vector and was found in all sites. Anopheles moucheti and An. nili were observed only in Nyabessang, which is surrounded by large rivers and dams offering suitable breeding sites for the development of larvae of these two species. Among the species of the An. gambiae complex, An. arabiensis was found in the two northern sites where the climate is drier than in the southern regions, which corresponds to reports from other sub-Saharan African countries where An. arabiensis was also found in drier areas [34, 35]. Furthermore, An. gambiae s.s. was predominant in Mangoum and Nyabessang while An. coluzzii represented the main species of the complex found in the other three sites. It is known that An. gambiae s.s. prefer larval habits with lots of sun exposure, while An. coluzzii are typically found in the man-made areas such as rice fields and more humid areas [36, 37]. The findings of this study corroborate with previously reported data [12, 18, 19], as Simatou and Gounougou are rice cultivation areas and Bonabéri is in the southern humid area. Similar results have also been reported from previous studies conducted in comparable eco-geographical areas in the country [38], though this study assessed vector bionomics over two consecutive years. Reviewing the trends over this period can help decision makers to assess and determine not only the tools that will be most effective, but also the optimal timing of deployment to achieve the desired impact. For example, the eco-geographical location of Nyabessang, surrounded by many rivers and dams with a high rainfall, favored the development of An. paludis and An. moucheti over leading An. gambiae s.l. at a specific period of the year. The high density of both species observed during the same period of the year could indicate the need for integrated strategies to control all species. On the other hand, Mangoum recorded the lowest species diversity with predominantly An. gambiae s.l. And few An. ziemanni recorded throughout both years of collections. This could be due to the location of the site and farming activities including corn and tomato gardening. Mangoum is a humid and sunny area located within a forest savannah, which is favourable for An. gambiae s.l. breeding sites, implying that vector control tools that target An. gambiae s.l. will likely be effective, such as any appropriate combination of ITNs.

The HBR of the Anopheles species varied seasonally at each site, which could be related to the eco-geographical location of the sites and peaked with either increasing rainfall and/or rice cultivation. Rainfall and rice paddies are known factors contributing to an increase in biting and consequently an increase in the malaria incidences in endemic countries [39, 40]. Biting in both Gounougou and Simatou peaked during the rainy season, which coincides with rice cultivation. This trend was observed over both survey years and highlights the need for malaria control strategy implementation, particularly during the peak transmission period which seems to recur from year to year. The NMCP in Cameroon has initiated seasonal malaria chemoprevention (SMC) in the northern regions [41] and provides free malaria treatment of children under five across the country. In addition, ITNs are distributed through mass campaigns and routine channels. However, it may be necessary to consider additional vector control measures such as complementary larval source management (LSM) or indoor residual spraying (IRS) where feasible. An impact evaluation may help determine if LSM could help control the diversity of Anopheles in areas where rice cultivation is conducted, such as in Simatou and Gounougou. In contrast to the northern sites, two biting peaks were recorded at the sites in southern part of the country, where two rainy seasons are observed yearly. Interestingly, all vectors showed slight (but not significantly different than indoor) outdoor biting patterns over both years and at all sites, even though several animal shelters were found in the northern sites, which could contribute to outdoor feeding of the vectors [42, 43]. Furthermore, the endophagic rates recorded in Gounougou and Simatou were lower than those of the southern sites because of the presence of a substantial number of cattle farms. However, HBRs were still high in Cameroon compared to some sub-Saharan African countries with similar geographical and climatic conditions [44, 45, 46, 47, 48].

Though malaria transmitted by An. gambiae s.l. was similar indoors and outdoors at all sites, the highest transmission was observed in Simatou, where six Anopheles species were found infected and a higher HBR was recorded compared to the other sites. This multiplicity of vectors could continue to worsen given recent reports on the potential transmission by sub-species of An. funestus s.l. in addition to those known in the An. gambiae complex [49, 50]. Though the diversity of malaria vectors has been described in the country [12, 17, 51], no specific vector control measures targeting the various species have been developed to date. All control efforts are channeled towards the main vector An. gambiae s.l., with the expectation that they will also have effects on the other vectors. However, the vectorial capacity of other vectors that are currently considered as secondary vectors needs to be closely monitored. Furthermore, this diversity of Anopheles vectors constitutes a cause for concern, considering that the current vector control interventions only target mostly endophagic and endophilic An. gambiae s.l. This strategy could alter vector dynamics, creating opportunities for niche partitioning and for other vectors to fill in the gap left by reduced populations of An. gambiae s.l. As observed in Nyabessang, An. moucheti and An. paludis yielded higher vectorial capacity and entomological inoculation rates compared to An. gambiae s.l. This may require deeper investigation into the ecology, transmission, and epidemiological impact of these vectors for targeted vector control interventions. Despite climate difference between the two northern sites, the pattern of the malaria transmission was similar, and all vectors observed had high parity rates suggesting they live long enough to transmit the disease, as the parity rates recorded at all sites were high for most of the vectors. This observation suggests that the current vector control tools implemented by the NMCP may have limited impact on the vectors. Cameroon recommended universal coverage and mass distribution of pyrethroid-only ITNs for a decade before introducing new types of nets during the 2022 mass ITN distribution campaigns. Even though some positive results were recorded on the decrease of morbidity due to malaria, more efforts are needed to reach elimination of the disease in the country. The use of ITNs was reported to be low among the target populations of Cameroon [52], indicating the need for social and behaviour communication programs to be undertaken by the NMCP.

Conclusion

Cameroon has a diverse and high density of Anopheles species, with An. gambiae s.l. as the main malaria vector in most geographical regions of the country. However, in this survey, An. moucheti and An. paludis were the main malaria vectors in Nyabessang. Seasonal variations of HBRs and the indoor resting density of An. gambiae s.l. were observed in all sites. Anopheles gambiae s.l. was observed to bite more indoors in Mangoum and more outdoors in Gounougou, Bonabéri, and Nyabessang, with biting occurring until the early morning hours at all sites.

Eleven Anopheles species and subspecies were involved in malaria transmission and An. gambiae s.l. highly contributed at all sites, except in Nyabessang, where An. moucheti and An. paludis accounted for 25.9% and 32.1% of EIR, respectively. This study highlights the urgent need for integrated vector control interventions considering all potential vectors to reduce malaria transmission and burden in Cameroon. Based on these results, a literature review, and the insecticide resistance monitoring data recorded across the country, the NMCP conducted the 2022 ITN mass distribution using PBO ITNs and dual active ingredient Interceptor G2 ITNs in regions with high malaria endemicity, high transmission, and high insecticide resistance intensity. The data could also support the deployment of IRS in targeted sites with timing determined by the trends observed over the collection years. Targeted LSM could be an additional option to reduce the peak biting and transmission in northern areas, where rice cultivation increases the mosquito population density, but should be implemented in the context of an impact evaluation.