The “Cancer Bush” Sutherlandia (syn. Lessertia) frutescens. An Example of Promotional Research, or Is There Scientific Merit?

Keywords

anticancer - cytotoxic - Fabaceae - herbal medicine - marketing - Sutherlandia frutescens - traditional remedy

Introduction

Sutherlandia (syn. Lessertia) frutescens (L.) R.Br. ex W. T.Aiton (Fabaceae), colloquially known as the ‘cancer bush’ or ‘Kankerbos’ (Afrikaans), is a popular herbal remedy in South Africa, revered for its medicinal properties and, as the common name implies, for its purported anticarcinogenic properties. This drought-resistant shrub belongs to the Fabaceae family, the third-largest family of flowering plants, and is widely distributed throughout the drier regions in the Western, Eastern, and Northern Cape and parts of the KwaZulu-Natal provinces of South Africa, as well as other parts of southern Africa, including Botswana, Zimbabwe, and Namibia [1]. Despite the widespread use and commercial availability of S. frutescens, concrete scientific data surrounding its mechanism of action against the proliferation of carcinomic cells remains limited, and the evidence for its efficacy as a prophylaxis and/or treatment for cancer is seemingly inconclusive [2]. Due to these uncertainties, it was decided to collate all available scientific data regarding the antiproliferative effects of the genera Sutherlandia and Lessertia and critically evaluate the presented data to ascertain if there is any scientific merit to the anticancer health claims.

An updated overview of this very popular herb is provided here. This review is divided into three sections: the first addresses general topics such as botany and traditional use, and the second briefly covers the literature on its phytochemistry and toxicity, whilst the bulk of this review covers its antiproliferative properties and a critical appraisal of published data. The objective of this review was therefore to collate all scientific data on the genera Lessertia and Sutherlandia with the main focus, as the vernacular name ‘cancer bush’ implies, being their anticarcinogenic, antiproliferative, antineoplastic, and cytotoxic activity and to critically evaluate the data against established guidelines. Recommendations regarding the use of this popular herb, as well as to identify a direction of future research, are also presented.

Search Strategy

Scientific literature published in English up to December 2024 was retrieved from scientific databases (Science Direct, Web of Science, Scopus, PubMed, and Google Scholar), as well as from several (historical) books regarding medicinal plants and ethnobotany. As an indication of the overall amount of all scientific research that has been conducted on these two genera, the Web of Science lists only 55 publications using the keyword ‘Lessertia’ and 201 publications using the keyword ‘Sutherlandia’, with the vast majority not applicable to this review.

Botanical Aspects

Sutherlandia frutescens is native, but not limited to, the Cape Floristic Kingdom and is specifically found in the Fynbos biome, which hosts one of the greatest varieties of plant species in the world. Classified in the class Magnoliopsida, order Fabales, and genus Sutherlandia, this plant has been suggested for reclassification under the genus Lessertia, based on the suggestion that it has adapted to bird pollination [1]. However, the name Sutherlandia is commonly used on commercial products and colloquially and hence will be used to describe this genus throughout this manuscript. According to the International Plant Name Index, the following Sutherlandia species are recognised: S. darumbium Bertero; S. floribunda Carrière; S. foliolata Lange; S. frutescens; S. humilis E.Phillips & R. A.Dyer; S. incana E.Mey.; S. littoralis J. F.Gmel.; S. microphylla Burch. ex DC.; S. montana E.Phillips & R. A.Dyer; S. speciosa E.Phillips & R. A.Dyer; S. tomentosa Spreng.; and S. vesicaria Spreng.

Sutherlandia frutescens is a medium-sized shrub (roughly 0.2 – 1.5 m in height) with fine silver-green leaves and red butterfly-shaped flowers, while its seedpods are large, balloon-like, and slightly reddish, hence the common name ‘balloon pea’ ([Fig. 1]) [1], [3]. Although mostly harvested from the wild, it is also cultivated in community gardens and commercially. The plant is known by over 25 different common names, many reflecting its characteristics, such as ‘kalkoenbos’ (turkey bush) and others referring to its medicinal uses, like ‘Kankerbos’ (cancer bush), ‘insiswa’ (dispels darkness), and ‘lerumo lamadi’ (spear of the blood) [1], [3]. The taxonomy is debated, with various authors classifying the genus differently. The taxonomy of this genus was reviewed, and for the current review and the sake of thoroughness, both genera Lessertia and Sutherlandia and their purported effect on cancer will be detailed [4].

Fig. 1 a Sutherlandia frutescens inflorescences b leaves and flowers c seed pods and d Sutherlandia frutescens shrub (Photos: with permission of Random Harvest Nursery).

The Ethnopharmacology of Lessertia and Sutherlandia

The first records of its use date back to the Dutch colonists in the Cape, who likely learned of it through interactions with the indigenous Khoi-San people. The Khoi-San and Nama traditionally utilised decoctions of Sutherlandia to cleanse wounds and reduce fevers [1]. One of the earliest mentions of S. frutescens and its use as a possible cancer treatment was documented in 1895, when it was noted that it was commonly used for the treatment of cancer, but a physician concluded that it was essentially ineffective [5]. A valuable source of information regarding the traditional use of plants in southern Africa was published in 1932 by Watt and Breyer-Brandwijk, with a second edition published in 1962 [6]. This resource is not electronically available, and there is a significant risk that this valuable and insightful source may eventually be lost. Therefore, information from this resource should be collated and made available electronically via scientific publications. Here we present all that is documented regarding the genus Lessertia and Sutherlandia as recorded by [6].

Lessertia annularis Burch. was mentioned as the possible cause of cotyledonosis (‘krimpsiekte’) but this was later attributed to Cotyledon ventricose Burm. However, L. annularis was stated to be toxic. L. brachystachya DC. was shown to be toxic to cattle during feeding experiments, whereas L. physodes E.&Z. did not show any toxicity. L. argentea Harv. was utilised as an eye lotion but also to alleviate flatulence and colic, and L. tomentosa DC. was documented to be used by the early settlers as a remedy for ophthalmic conditions. An infusion of S. frutescens R.Br. var. incana E.Mey. was used orally as a cough remedy and for cancer and externally as a female hygiene product. It is interesting to note that the infusion induced vomiting when it was prepared too ‘strong’, but no details regarding the preparation method were provided. Similarly, a decoction of S. humilis Phill. & Dyer and an infusion of the bark and leaves were used as a treatment for cancer as well as influenza.

Sutherlandia microphyla Burch. was stated to be very popular and the most commonly used Sutherlandia species throughout South Africa. It was specified that a ‘weak’ infusion or decoction of the leaves was taken in unlimited amounts to treat a wide variety of ailments, including stomach complaints, internal cancers, uterine complaints, influenza, liver disease, rheumatism, inflammations, haemorrhoids, dropsy, and backache. An infusion was also consumed for amenorrhea and as a cancer prophylactic. The powdered leaf mixed with syrup was claimed to be soothing for a troublesome cough and was also used as a tonic to improve appetite and digestion. However, when the infusion was prepared too ‘strong’, without defining the preparation method, it caused sweating and acted as a purgative.

The Southern Sotho used the plant as an infusion for the treatment of fever as well as ‘dropsy of the heart’, whereas the Nama used a decoction to treat wounds, fevers, consumption (TB), chicken-pox, and a variety of other conditions. In North Africa, the leaf was used as a tonic and for indigestion and dysentery. A clinical trial was conducted using an infusion of the bark and the whole plant for the treatment of cancer, but the results of the trials were negative. The original document published in 1918, however, only stated that “Cancer bush (Sutherlandia frutescens). A half shrub, two to four feet in height, with scarlet flowers. It is supposed to cure cancer. Clinical experiments have proved this belief to be entirely without foundation” [6], [7].

It is claimed that the consumption of half a cup of Sutherlandia infusion or decoction three times daily would aid in the treatment of cancer and that the intake of a Sutherlandia decoction would prevent cancer metastasis [8], [9]. Improved quality of life of patients suffering from pancreatic, breast, and other cancer was reported with the administration of a Sutherlandia tonic [10], whereas it was asserted that a correlation exists between the use of Sutherlandia and a reduction in cancer-induced muscle wasting [11]. The traditional use of S. frutescens is categorised under wound healing, anticancer, antiviral, antibacterial, and anti-inflammatory treatments, along with the alleviation of gastrointestinal-related ailments and mood disorders (such as anxiety, depression, and mood imbalances). Generally, Sutherlandia is used as a bitter tonic, tranquiliser, and blood purifier [3].

In South Africa, S. frutescens is widely used by various healers, including herbalists (inyanga), diviners (isangoma), bush doctors (bossiedokters), Rastafarians, and both alternative and allopathic medicine practitioners, as well as by the public [1]. While Sutherlandia has long been appreciated as a decorative garden plant for flower arrangements, it is most renowned for its medicinal properties, particularly as an herbal medicine to improve the quality of life in HIV/AIDS patients [1], [12]. Several commercial farms currently produce S. frutescens to satisfy the growing demand for this popular herb, which is mainly being sold as an over-the-counter complementary medicine [1].

Phytochemistry

No comprehensive phytochemical analysis on this species has yet been conducted. The identified phytochemicals have been reviewed and include free amino acids, non-protein amino acids (L-canavanine, GABA), the cyclitol, pinitol, flavonoids, flavonoid glycosides, and cycloartane glycosides [4]. The flavonoid glycosides were identified as sutherlandins A to D [13] and the cycloartane glycosides as sutherlandiosides A – D, with sutherlandioside B commonly known as SU1 [14], [15]. The sutherlandins and sutherlandiosides are currently considered chemical markers for the species, and various publications could be found using these phytochemicals as quality-control marker compounds [3], [16], [17], [18]. Recently, a somewhat more in-depth analysis regarding the presence of other phytochemicals was conducted, and seven new cycloartane glycosides (sutherlandiosides E – K) and one oleanan-type saponin were identified [19]. Little information exists as to the antiproliferative activity of these compounds, and this aspect will be discussed in section 4.2.

Pharmacological Aspects

Toxicity

No serious adverse events have been reported with the use of S. frutescens, and only mild side effects such as dry mouth, diarrhoea, mild diuresis, and dizziness were described [11]. In 1908, it was stated that typical (effective) dosages for traditional infusions ranged from 2.5 g to 5.0 g of dried plant material daily [20]. High doses, such as a decoction prepared from 5 g taken daily for over 6 years, did not result in any reported side effects [4]. Toxicity studies on Sutherlandia have indicated a median lethal dose (LD₅₀) of 1280 ± 71 mg/kg in mice, suggesting the plant is relatively safe in mammals [21]. For commercial preparations, 600 mg of dried leaves per day is generally recommended, with precautions during pregnancy or lactation. In a safety study in vervet monkeys, doses up to nine times the recommended amount showed no significant toxic effects [22]. A clinical trial in 25 healthy adults found that 800 mg of leaf powder daily was well tolerated, with no serious side effects reported [23]. Some minor physiological changes were observed but remained within normal ranges. However, caution is advised, as Sutherlandia may interact with antiretroviral medications, diabetes drugs, and tuberculosis therapies [3], [11], [24].

Antiproliferative Studies

[Tables 1] and [2] provide, to the best of our knowledge, the details of all published antiproliferative studies conducted using S. frutescens. The source of plant material, plant parts used, and sample preparation methods are presented in [Table 1], whereas [Table 2] provides the cancer cell lines used, expression of activity (apoptosis, % inhibition, and IC₅₀), normal cell lines used, selectivity index, and IC₅₀ of positive controls.

Table 1 Plant origin, plant parts used, and extraction methods used in previous studies.

Plant source (location)	Plant parts used	Extraction method	Ref
Commercial source 300 mg tablet plus inert excipients	Unknown	Tablet extracted with 2.2 mL 70% ethanol	[25]
Collected at Kirstenbosch (WCP), Free State (FSP), and Northern-Province (NP)	Leaves, stems, flowers	10 g of powder extracted in 1 L boiling water. 40 mg/mL stock solution	[26]
Gift from Dr. C. Albrecht, Medical Research Council (SA)	Leaves	1 g of dried leaves boiled in 10 mL water for 15 min	[27]
Murraysburg (SA) S. frutescens (L.) R.Br. var. microphylla	Leaves and small twigs	1 g extracted in 10 mL 70% ethanol for 24 h, stock solution (100 mg/mL)	[28]
Murraysburg (SA) S. frutescens (L.) R.Br. var. microphylla	Leaves and small twigs	0.5 g in 10 mL boiling water and left overnight	[29]
Two S. frutescens collected at Colesberg (A) and Platvlei (B) and S. tomentosa (C) collected at Stilbay	Leaves and stems	70% ethanol for 24 h and stock solutions in water (100 mg/mL)	[42]
S. frutescens subsp. microphylla	Leaves and twigs	0.5 g in 10 mL boiling water and left overnight	[30]
Location unknown–in ocean. Collected from same plant 1 year apart.	Leaves and small twigs	0.5 g in 10 mL boiling water and left for 15 h	[31]
PhytoNova (Cape Town, South Africa) in tablet form containing 300 mg of raw herb powder compounded with inert excipients.	Unknown	Tablet extracted in 2.2 mL of 70% ethanol at room temperature for 2 h on a shaker. Supernatant diluted 1/50 and 1/100	[32]
Big Tree Nutraceutical and Thebe Natural Medicines (Cape Town, South Africa)	Unknown	Extracted with methanol, dried, and redissolved in ethanol at 50 mg/mL	[33]
Sutherlandia SU1 tablets	Unknown	3 g S. frutescens in 30 mL of 70% ethanol, under constant agitation for 24 h.	[34]
S. frutescens R. BR. (variety Incana E. Mey.) purchased from Big Tree Health Products (Fish Hoek, South Africa)	Leaves	1 kg of the powdered preparation was soaked in boiling water and left overnight for extraction.	[35]
Raw powdered plant material of S. frutescens purchased from Afrinatural holdings (KwaZulu-Natal, South Africa).	Unknown	1 : 10 ratio in water (plant : liquid)–10 g raw powder in 100 mL distilled water and the suspension was sonicated at 45 °C for 45 min. A stock solution of 8 mg/mL was prepared, and subsequent concentrations were prepared ranging from 1 to 8 mg/mL.	[37]
Collected from 7 locations throughout SA	Leaves	1 g leaves extracted with 20 mL of 75% ethanol. Ultra-sonicate for 60 min. Stock extract of 1 mg/mL	[36]
Dried and milled, purchased from Afrinatural holdings	Unknown	1 : 10 ratio in water (plant: liquid) followed by 45 min sonication at 45 °C. 1 – 8 mg/mL in culture medium stock solutions	[38]
Dried and milled, purchased from Afrinatural holdings	Unknown	1 : 10 ratio in water (plant: liquid) followed by 45 min sonication at 45 °C. 1 – 8 mg/mL in culture medium stock solutions	[38]
Sutherlandia frutescens leaves purchased from Parceval Pharmaceutical Industry	Leaves	Macerated in MeOH (3 × 24 h) at room temperature	[19]

Table 2 Cell lines used, cytotoxicity, selectivity index, IC₅₀, and positive controls of all studies testing S. frutescens.

nr	Cell lines used	[Apoptosis]	% inhibition	IC50	Normal cell lines (IC50)	Selectivity index	Positive control (IC50)	Ref
* Activity expressed as a dilution factor
1	MCF-7, MDA-MB-468, Jurkat, HL60	ND	ND	IC50 ± 1/250* (MCF7), 1/200 (MDA-MB-468), 1/150 (Jurkat) 1/200 (HL60)	ND	ND	Camptothecin 1 uM (348 ng/mL), paclitaxel 10 nM (8.54 ng/mL)	[25]
2	CHO, Caski, Jurkat T lymphoma	Observed at 3.5 mg/mL (WCP), > 10 mg/mL (FSP and NP)	ND	ND	ND	ND	ND	[26]
3	DU-145, MCF-7, MDA-MB-231	Minor inhibition at 50 ug/mL against DU-145 and MCF7 and stimulation of MCF-12A and MDA-MB-231	ND	ND	MCF-12A	ND	Cisplatin 0.27 µg/ml (MCF-7), 0.14 µg/ml (MCF-12A)	[27]
4	MCF-7	70% ethanolic extracts induced morphological hallmark in MCF-7	40 – 50% inhibition after 72 h with 1.5 mg/mL plant material equivalents		ND	ND	No positive control (vehicle-treated controls)	[28]
5	MCF-7 and MCF-12A	ND	MCF-7 (26% at 10 mg/mL)		MCF-12A (49% at 10 mg/mL)	Claims selectivity but unclear both cell lines significantly inhibited–no IC50s provided	Unclear–no drug positive control. Vehicle controls	[29]
6	SNO	Extract A: 2.5 mg/mL: early apoptosis, 36% of cells and late apoptosis, 13% of cells. 5 mg/mL: early apoptosis 22% and late apoptosis 21%. Extract B: Early (25%) and late (64%) apoptosis at 2.5 mg/mL. Early (14%) and late (74%) apoptosis at 5 mg/mL. Extract C: Early (30 – 32%) and late (31 – 32%) apoptosis at 2.5 – 5 mg/mL.	ND	ND	Cell viability reduction after 24 h: Extract A (2.5 and 5 mg/mL). B (0.5 – 10 mg/mL). C (1 – 10 mg/mL). After 48 h: Extract A (2.5 and 5 mg/mL). B (0.5 – 10 mg/mL). C (0.5 – 10 mg/mL).	ND	ND	[42]
7	MCF-7 and MCF-12A	ND	1 mg/mL at 48 h exposure GI50 parameters calculated for the MCF-7 (51.79% of control)	ND	MCF-12A (93.22% of control).	ND	No positive control (vehicle-treated controls)	[30]
8	MCF-7 and MCF-12A	MCF-7 (1/10, 20.59%),	ND	ND	MCF-12A (4.9% and 1.6%)	ND	Actinomycin D (0.1 µg/ml) and tamoxifen (20 µM)	[31]
9	CaCo2	Increased modulation of apoptosis-related proteins (PARP)	1/100 (no significant effect), 1/50 (48.83%)	ND	ND	ND	No positive control	[32]
10	Human (PC3 and LNCaP), Mouse (TRAMP-C2)	ND	ND	IC50 of 167 µg/mL in PC3, 200 µg/mL in LNCaP 100 µg/mL in TRAMP-C2	ND	100 µg/mL in TRAMP-C2	No drug controls	[33]
11	A375, Colo-800 melanoma; HDFα and Hek 293 normal cell lines	Stimulation of intrinsic and extrinsic caspase-dependent apoptosis	ND	A375 (625 µg/mL) but control cell more sensitive illustrating non-selectivity at these high doses.	HDFα (300 µg/mL) Hek 293 (625 µg/mL 90% survival)	ND	Unknown	[34]
12	SKNBE(2) and SHSY5Y, KMST6	ND	ND	SKNBE (2) (2.3 mg/mL) and SHSY5Y (2.2 mg/mL) KMST6 at 5 mg/ml concentration cell survival was 50.9%.	ND	ND	ND	[35]
13	3D colorectal cell model (LS180)	ND	ND	LS180 (2.625 mg/mL)	ND	ND	Paclitaxel (94.595 nM)	[37]
14	DLD-1	ND	ND	4 samples > 200 µg/mL with Colesberg sample 158.7 ug/mL. No control			ND	[36]
15	LS180 cells and 2D cells	ND	ND	2630 µg/mL	ND	ND	Paclitaxel 0.081 µg/mL	[38]
16	LS180 cells 3D cells (spheroids)			Treated with 201 and 403 µg/mL	ND	ND	Treated with 0.006 µg/mL paclitaxel	[38]
17	MCF-7	ND	No 50% inhibition was achieved at 100 µg/mL for extract, fraction, and sutherlandiosides and 1 saponin	ND	ND	ND	ND	[19]

The study conducted by Tai appears to be the first in vitro investigation on this topic and comprised the use of commercial Sutherlandia tablets [25]. A 70% ethanol extract used during this study was tested against the human mammary adenocarcinoma (MCF-7; MDA-MB-468), human leukaemia (Jurkat), and human promyelocyte (HL60) cell lines. The reported 50% inhibition (IC₅₀) was expressed as a dilution factor of the primary 70% ethanol extract at the following dilutions: 1/250 for MCF-7 (most active), 1/200 for MDA-MB-468, 1/200 for HL60, and 1/150 for Jurkat (least active). The positive controls, camptothecin and paclitaxel, yielded IC₅₀ values of 1 µM (348 ng/mL) and 10 µM (8.54 ng/mL), respectively.

The leaves, stems, and flowers of Sutherlandia were tested against Chinese hamster ovary (CHO), Caski (cervical carcinoma), and Jurkat T lymphoma cell lines, where apoptosis was noted at 3.5 mg/mL for plants collected in the Western Cape Province (WCP) and approximately 10 mg/mL for plants originating from the Northern Province (NP) and Orange Free State (OFT) [26].

The cytotoxicity of an aqueous Sutherlandia extract was tested against prostate cancer (DU-145), breast cancer (MDA-MB-231 and MCF-7), and non-malignant breast (MCF-12A) cell lines. The extract yielded IC₅₀ values of > 50 µg/mL, which exceeded the limit of activity for crude extracts (30 µg/mL) as set by the American National Cancer Institute. Interestingly, the positive control, cisplatin, yielded respective IC₅₀ values of 0.27 µg/mL (MCF-7) and 0.14 µg/mL (MCF-12A), indicating a lack of selectivity [27].

Crude extracts of S. frutescens were tested for activity against the human breast adenocarcinoma (MCF-7) cell line, and an IC₅₀ value of 1.5 mg/mL was reported after 24 hours of exposure [28]. In a follow-up study of the cytotoxic effects of a crude S. frutescens extract on the MCF-7 and MCF-12A cell lines, a dose-dependent inhibition of 26% viable cells (MCF-7) and 49% (MCF-12A) was reported at a concentration of 10 mg/mL compared to a vehicle control. No positive drug controls were used in either of these studies [28], [29]. In another follow-up study, an aqueous extract of S. frutescens against MCF-7 and MCF-12A was investigated. Inhibition of roughly 50% and 7% was reported against the two cell lines, respectively, at a 1 mg/mL dose after 48 h exposure. No positive drug controls were used [30].

An aqueous extract of S. frutescens was tested for activity against MCF-7 and MCF-12A cell lines, and optimal apoptosis was shown for a 1/10 dilution. The MCF-7 cells revealed that 20.59% of cells treated indicated apoptosis, whereas only 2.02% of MCF-12A cells indicated apoptosis [31]. In another study, the cytotoxic effects of S. frutescens on a colon cancer cell line (CaCo2) were investigated, and a 1/50 dilution of the S. frutescens extract yielded a more significant decrease in cell viability compared to a 1/100 dilution [32].

The antiproliferative effects of S. frutescens in human (PC3 and LNCaP) and mouse (TRAMP-C2) cell lines were investigated, and IC₅₀ values at 167 µg/mL, 200 µg/mL and 100 µg/mL against the PC3, LNCaP and TRAMP-C2 cell lines, respectively, were reported [33].

The apoptotic effects of S. frutescens in melanoma cell lines (A375 and Colo-800), as well as in HDFα and Hek 293 normal cell lines, were tested. The results of this study yielded an IC₅₀ value for the A375 cell line of 0.625 mg/mL following 24 h of treatment, whereas the Colo-800 cell line showed higher resistance with 98% cell viability after 24 h of treatment. The HDFα normal cell line showed high sensitivity to S. frutescens treatment with an IC₅₀ value at 0.3 mg/mL. The Hek 293 normal cell line, however, showed the highest resistance to treatment with a slight proliferative effect seen at a 0.625 mg/mL treatment concentration between 24 and 48 hours of treatment [34].

The effects of S. frutescens were investigated on the cell viability of human malignant neuroblastoma cells (SKNBE(2) and SHSY5Y) along with non-cancerous fibroblast cells (KMST6) using an MTT (3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyltetrazolium bromide) cell viability assay and the clonogenic assay [35]. Moreover, flow cytometry was utilised to detect the accumulation of intracellular reactive oxygen species (ROS), as well as the loss of mitochondrial membrane potential (MMP). Caspase-9 activation was also used to detect cell apoptosis. IC₅₀ values obtained were 2.3 mg/mL and 2.2 mg/mL for the SKNBE(2) and SHSY5Y, respectively, whereas the KMST6 were less sensitive with an IC₅₀ of 5 mg/mL.

The bioactivity of S. frutescens extracts from different regions via in vitro antioxidant and antiproliferative effects against the colon adenocarcinoma cell line (DLD-1) was investigated [36]. The results of the antiproliferative assay revealed that all extracts produced cytotoxic effects at the highest treatment concentration (200 µg/mL), and the most potent antiproliferative activity was shown by the plant extracts originating from Colesberg, Zastron, and Gansbaai 1 with respective IC₅₀ values of 158.7 µg/mL, 172.7 µg/mL, and 176.7 µg/mL.

Two-dimensional and three-dimensional spheroid LS180 colorectal cancer cells were used to test the activity of S. frutescens extracts. Commercial S. frutescens and Xysmalobium undulatum (L.) W. T.Aiton (Apocynaceae) were first tested against LS180 colorectal 2-D cancer cells, and paclitaxel was used as positive control. IC₅₀ values of 2.63 mg/mL and 0.098 mg/mL were reported for S. frutescens and X. undulatum, respectively, as compared to the positive control, which afforded an IC₅₀ value of 0.081 µg/mL (S. frutescens were therefore 32,469 times less active than the positive control in 2D assays). Half the IC₅₀ concentration (1.315 mg/mL) of S. frutescens was used to test a newly established 3D spheroid model. They evaluated soluble protein content, glucose consumption, intracellular ATP, and extracellular adenylate kinase (AK) release and compared these parameters with the positive control, paclitaxel. The authors conclude that “S. frutescens and X. undulatum crude aqueous extracts both showed notable cytotoxic effects and potential anticancer activity relative to the known chemotherapeutic drug paclitaxel, with decreased growth, cell viability, and glucose consumption following treatment” [37], [38].

The most recent study on Sutherlandia extracts was conducted by [19] who identified 8 new compounds and tested the extract, enriched fraction, and 10 of the 12 purified compounds against an MCF-7 cell line. No adequate inhibition at the highest concentration of 100 µg/mL tested was observed.

Critical Evaluation

There are several minor but also major concerns that need to be discussed. The minor concerns will first be detailed, followed by a more in-depth discussion of the major concerns. The study conducted by [25] does not provide a precise IC₅₀ value ([Table 2]), even though it is relatively uncomplicated to determine. However, when the IC₅₀ value of the positive control (8.54 ng/mL) is considered and an extraction yield of 10% is assumed, then the IC₅₀ value against MCF-7 can be calculated as 54 µg/mL (roughly 6300 times less active than the positive control). This is an estimate with the authors stating that the IC₅₀ “is about 1/250′′ dilution. Moreover, for the MCF-7 cell line, the positive control as indicated in Fig. 3 of [25] clearly inhibits the cells far below its IC₅₀ value with only about 10% of cells surviving at the concentration tested. The IC₅₀ is therefore likely to be significantly lower than reported and, subsequently, the difference between the S. frutescens sample and positive control much larger than reported.

In the study conducted by [26], it is unclear where the plant material was collected ([Table 1]). It mentions Kirstenbosch, Western Cape Province, “Bloiemfontent”, “Free State”, and Northern Province, but it is not specified what these locations refer to as “Bloemfontein” is the capital city of the Free State province in South Africa and no Northern Province exist (the North-West and Northern Cape provinces do exist). Apoptosis was observed in one sample at a concentration of 3.5 mg/mL, but no apoptosis was detected in the other two samples ([Table 2]). No positive control or non-cancerous cell lines were used, and hence, no selectivity index was determined. The conclusion states: “In summary, the claim by traditional healers that Sutherlandia frutescens has anticancer properties has been partially validated by identifying an extract in the plant that has marked apoptotic activity”. We tend to disagree with this conclusion when the extremely high concentrations tested (3.5 mg/mL) are considered.

The findings highlighted by the study conducted by [27] do not specify the source of the plant material, only stating that it was a gift from the Medical Research Council of South Africa. The study does not provide a specific IC₅₀ value (rather % inhibition), and S. frutescens performed poorly compared to five other plant species tested. The authors included positive controls and used acceptable guidelines to conclude that S. frutescens did not show acceptable activity and therefore continued work on other plant species.

The source of plant material was clearly defined by the study conducted by [28]; however, the study did not include positive controls or non-cancerous cell lines. They expressed the IC₅₀ as 1.5 mg dry plant material/mL; therefore, no exact IC₅₀ was provided. Given an extraction efficiency of 10%, the actual IC₅₀ is likely around 10 times lower. In a very similar study, [29] once again clearly defined their source of plant material but prepared an aqueous extract instead of a 70% ethanol extract. Again, no positive controls were used but a non-cancerous MCF-12A was included. At a high concentration of 10 mg/mL, some selectivity was claimed with the plant extract leading to a survival of only 25% of the MCF-7 cells vs. 54% of the MCF-12A cell line. However, from Fig. 1 in [29], it is clear that the reverse seems to occur at 5 mg/mL, with an estimated 64% survival of MCF-7 vs. 56% against MCF-12A. The paper states: “The non-tumorigenic MCF-12A cell line is less susceptible to apoptosis induction at higher concentrations of S. frutescens compared to the tumorigenic MCF-7 cell line.” The observed selectivity at 10 mg/mL vs. 5 mg/mL is, in our view, more a function of the high dosage, which will kill the cell (and will likely kill any cell type) with the slight ‘selectivity’ only observed under certain conditions due to inherent differences in cell types used–therefore, no true selectivity.

In a follow-up study conducted by [30], the same cell lines were used (MCF-7 and MCF-12A), and a dose of 1 mg/mL and 48 h of exposure was chosen based on a dose-response study. The authors do show that by using these dosing parameters, a certain level of selectivity is evident with the MCF-7 being more sensitive. It is, however, clear from Fig. 1 in [30] that, at different concentrations or exposure times, the exact opposite seems to occur with the MCF-12A being more sensitive than the MCF-7 cell line (e.g., at 2.5 mg/mL). Moreover, no positive controls were included, making these results hard to interpret.

In [Table 1] of [31], the source of the plant material is not clearly specified with the coordinates provided (35°51′32.12″S; 18°39′45.69″E) pointing to a location in the ocean. For sample preparation, the study used 10 mg/mL of plant material in water (rather than an extract concentration). Fig. 2 in [31] is unclear, making it difficult to understand how the differences could be considered “significant”. The significance is expressed compared to the growth medium (vehicle) control and does not include any statistically significant differences between the MCF-7 and MCF-12A cell lines. The manuscript mainly deals with the “differential cellular interaction” and in essence indicates that there may be a different mode of action against the two cell lines (apoptosis vs. autophagy) but no major difference in selectivity.

In the study conducted by [32], dilutions are utilised instead of specifying exact concentrations. The study lacks a positive control and does not provide an IC₅₀ value. The 1/100 dilution did not provide any significant activity, while the 1/50 dilution inhibited the Caco2 cells (48.83% survival). A 1/50 dilution equates to roughly 272 µg/mL when a 10% extraction yield is used in the calculations.

The study carried out by [33] included both in vitro and in vivo tests (although it lacked positive controls, and the plant parts used in this study remain unknown). The study also tested a fraction, mainly consisting of sutherlandioside D, and reported an IC₅₀ value of 1.8 µg/mL compared to 30 µg/mL for the crude extract against stimulated Shh Light II cells. It is unfortunate that sutherlandioside D was not also tested against the PC3, LNCaP, and TRAMP-C2 cell lines, making comparison of the results difficult. The in vivo mice experiments yielded some unforeseen results. Sutherlandia frutescens was mixed with mouse diets at three concentrations (0.05%, 0.25%, or 1% (wt/wt)) and fed to mice who developed autochthonous prostate tumours. No dose response could be observed, as with the in vitro experiments, and the 0.05% yielded the best results. However, it is unclear if the difference between the 0.05% treatment group and the control group is statistically significant (Fig. 7 in [33]–data not provided).

No positive controls were included in [34], and they reported a 50% inhibition at 625 µg/mL. However, the non-cancerous cell line HDFα afforded a roughly 50% inhibition at 300 µg/mL, indicating a lack of selectivity. The authors correctly concluded, “However, the relatively high concentration of the extract required to elicit such a response may pose a significant limitation for the clinical use of S. frutescens for the treatment of cancer.”

No positive controls were included in the study conducted by [35]. They do, however, show a selectivity index of roughly 2 between the SKNBE and KMST6 cell lines but at extremely high dosages (IC₅₀ of 2.3 vs. 5.0 mg/mL). One study was not included in the tables due to the difficulties in interpreting the results [39]. Sutherlandia frutescens, canavanine, GABA, and pinitol were tested for activity against SiHa cervical cells. From Fig. 1 in [39], it is abundantly clear that absolutely no dose-response curve can be observed. It is entirely unclear from the data presented how IC₅₀ values for all samples tested could be calculated. For the S. frutescens extracts tested at 50, 100, and 200 µg/mL, all concentrations afforded ~ 50% cell survival, whereas at 150 µg/mL, cell viability was presented as 250%. The same trend can be observed for the three compounds, with the same inhibition pattern observed at all concentrations tested.

Sutherlandia frutescens samples were collected from seven different geographic locations and tested against DLD-1 colon cancer cells [36]. Unfortunately, no positive or vehicle controls were used. The best activity of 158.7 µg/mL was reported for one specific sample with four samples presenting IC₅₀ values of > 200 µg/mL. This indicated that chemotypic variation may affect the bioactivity, but in the absence of controls, it remains difficult to form any firm conclusions. The findings reported by [38] reported activity of S. frutescens extracts that are ~ 32 500 times less active than the positive control in 2D assays. The positive control paclitaxel afforded an IC₅₀ value of 0.081 µg/mL, and if this is extrapolated with an average chemotherapeutic dose of 135 and 175 mg/m² for paclitaxel, it will equate to an S. frutescens dose of 4.39 – 5.69 kg/m2, which is clinically irrelevant. However, the study concludes that “the results of this study suggest that the anticancer activity of the S. frutescens crude aqueous extract, specifically in colon cancer, may even surpass that of the model chemotherapeutic drug, paclitaxel”. We simply disagree with this statement, and this brings us to a discussion of the major issues in many of the publications that need to be highlighted.

There are three main and, in our view, serious concerns. First, the absence of proper controls, especially positive controls. Positive controls are crucial for several reasons. First, bioassays involving living cells or organisms, both in vitro and in vivo, can be challenging and prone to significant variability in pharmacological response. Thus, incorporating a positive control is essential to confirm that the test system is functioning correctly. A negative control, such as a solvent or carrier, can also fulfil this role to some degree, and ideally, both a positive and negative control should be used [40]. Second, a positive control provides a benchmark for comparison, allowing researchers to evaluate whether results, such as the IC₅₀ of the positive control, are consistent with previous findings or literature and how the test sample compares to the positive control. Many academic journals (such as the Journal of Ethnopharmacology, Phytomedicine Plus, Planta Medica, etc.) emphasise the importance of positive controls, and it is stated in the author guidelines that the absence of the use of proper controls (e.g., recognised positive controls) may lead to immediate rejection of a manuscript [41]. This policy underscores the essential role that positive controls play in maintaining the integrity of scientific research. Without a positive control, it becomes impossible to validate the assayʼs performance, raising doubts about the accuracy of the results and the robustness of the experimental design [42]. Furthermore, positive controls provide a frame of reference that ensures consistency across scientific studies. For example, researchers can compare the activity of a test sample (such as S. frutescens) to that of a well-established active drug (such as paclitaxel), facilitating meaningful comparison and interpretations of the findings. By adhering to this standard, researchers not only increase the credibility of their work but also contribute to the overall advancement of science, preventing the publication of misleading or inconclusive data.

Only five of the studies ([Table 2]) made use of positive controls [25], [27], [31], [37], [38]. However, many studies failed to include positive controls or only used solvents as controls, making their findings far less useful and reliable [26], [28], [29], [30], [32], [33], [34], [35], [36], [43]. It was stated that “If no positive control was used in a standard assay, it is impossible to interpret the resulting data with regard to their scientific content” [44]. The importance of positive controls can also be explained by precision and accuracy. An increase in sample replicates will lead to a more precise result but cannot provide accuracy. Only a positive control can provide accuracy; hence, the lack of a positive control may lead to very precise but completely inaccurate results.

The other main concerns are the exceptionally high doses tested and the lack of selectivity studies, which, in essence, is the holy grail of drug discovery in the field of cancer research. Regarding high doses, the National Cancer Institute of the United States uses the following guidelines regarding plant extracts: IC₅₀ < 10 µg/mL, highly active; 10 – 50 µg/mL, moderately active; 50 – 200 µg/mL, weak activity; and > 200 µg/mL, inactive [45]. It was also suggested that an IC₅₀ < 100 µg/mL for extracts and < 25 µM for compounds should be considered the minimum acceptable level of activity [46], while Gertsch recommends < 50 µg/mL and < 5 µM, respectively [47]. Here, Paracelsus’ dictum, “The dose makes the poison”, which emphasises the importance of dose in determining efficacy and safety, can be highlighted [46], [47].

As an example, the study done by [29] can be discussed. Other than the lack of a positive control, the dosage used was 10 and 5 mg/mL (at a 10% extraction yield, it will equate to a dosage of 1.0 and 0.5 mg/mL). These high doses will likely kill any cell type, which was eloquently illustrated by the different responses observed between the cancerous MCF-7 and non-cancerous MCF-12A at 10 mg/mL, where MCF-7 was more susceptible, but at 5 mg/mL, the MCF-12A was more susceptible. Moreover, even if an IC₅₀ of 1 mg/mL is assumed and this value is compared to a positive control, paclitaxel, as published by [25] against MCF-7 of 8.54 ng/mL, a ~ 120 000-fold difference in IC₅₀ values is observed. When this is translated into an in vivo chemotherapeutic dose (135 – 170 mg/m² for paclitaxel) many kilograms/m² of S. frutescens must be administered, which is simply not possible. Other high-dose experiments, such as [26] who reported inhibition at 3.5 mg/mL, [43] at concentrations of 2.5 – 5 mg/mL, [35] who tested at 2.3 mg/mL, [37] at 2.625 mg/mL, and [38] at doses of 2.63 mg/mL, are highly concerning, as these high doses are clinically irrelevant. These have been called “outrageously high concentrations”, adding that “…claims of activities at absurd concentrations are extrapolated to the human situation…” [44]. We tend to agree and must conclude that most of the studies presented here were tested at far too high concentrations to make the studies scientifically meaningful.

The selectivity index (SI) is another crucial aspect and is described as the ratio between the toxic concentration of a sample relative to its bioactive concentration. Furthermore, an ideal drug or active compound would possess a high toxic concentration directed towards malignant cells while maintaining a low bioactive concentration [48], [49]. Determining the SI during any research surrounding herbal medicine or isolated active compounds is necessary to evaluate if further studies are warranted [49]. Several studies investigated the selectivity of S. frutescens extracts between the tumorigenic breast cancer cell line MCF-7 and the non-tumorigenic MCF-12A cell line [27], [29], [30], [31]. The ideal situation would be that the MCF-7 cell line would be highly sensitive to S. frutescens, whereas the MCF-12A cell line would not be sensitive at all, resulting in a high SI. In essence, the higher the SI, the better. Unfortunately, only two studies included the positive controls, cisplatin [27] and tamoxifen [31], which places a question mark on the validity of the other studies. Moreover, excessively high doses were selectively tested, and in addition, no SI values were presented even though the study designs are essentially focused on calculating an SI value. It was shown that at 50 µg/mL, little to no activity was observed against the MCF-7 cell line, whereas some stimulation was observed in the MCF-12A cell line [27]. It is of interest to note that the positive control also did not show any selectivity as the IC₅₀ of cisplatin was given as 0.27 µg/mL and 0.14 µg/mL against the MCF-7 and MCF-12A cell lines, respectively. Due to the low activity of S. frutescens, no IC₅₀ value was provided and, hence, no SI was presented with the authors correctly discontinuing further work on S. frutescens based on the observed data.

The data presented by [29] was difficult to interpret. The MCF-12A cell line seems to be more sensitive to S. frutescens extracts at 3 and 5 mg/mL, whereas the MCF-7 cell line is slightly more sensitive at 10 mg/mL. Based on this, a decision was made to continue more advanced studies at the higher dosages, such as flow cytometry and electron microscopy, whilst the decision should have been that there was no clear selectivity at these very high dosages, and the focus and resources should have been spent on more promising candidates.

In the study by [30], the same pattern emerges. At very specific (high) concentrations and treatment times, MCF-7 seems to be more sensitive, but at different concentrations and treatment times, the opposite appears to happen. Here again, more advanced studies were conducted at a very specific concentration and exposure time, where there seems to be a slight difference in favour of MCF-7, whereas an unbiased opinion would have been that the quality of data was not good enough, and scarce resources should have been spent on more promising candidates.

Of the 17 studies in [Table 2], only one discussed an SI [33]. The remaining studies did not mention any calculated SI value even though they alluded to the presence of selectivity. Evaluating the antiproliferative activity of a sample using only malignant cell lines, without considering the SI, provides a poor indication for further clinical studies [50]. Optimally, with the provision of an SI, a drug should be able to inhibit cancer cells at a very low concentration while normal cells should not be affected during in vitro scientific studies [51]. Hence, without the SI data for the reported bioactive samples, the value of such publications is significantly limited [52].

Limitations of In Vitro Models

In vitro models are used because they are relatively inexpensive and easy to perform and many samples can be tested simultaneously, making it an ideal tool in large drug-discovery projects. Moreover, there are fewer ethical concerns with in vitro studies than with in vivo animal studies. It essentially acts as a funnel to filter out the inactive or less active samples in order to focus on the most promising leads, which can then be further studied in much more expensive in vivo models [53]. It is essential to recognise that laboratory-based cytotoxicity tests (in vitro) do not consider important pharmacokinetic factors like absorption, metabolism, and excretion, which are critical in actual clinical trials (in vivo) [54]. Consequently, the outcomes of in vitro cytotoxicity studies should be viewed as an approximate guideline rather than direct indicators of in vivo cytotoxicity [54]. The old adage “In vitro simplicitas, in vivo veritas” (in vitro simplicity, in vivo the truth), as stated by [44], should here be considered. It is therefore unfortunate that no in vivo studies, other than the small-scale mice experiments [33], have been conducted, and yet major claims regarding the anticancer activity of S. frutescens are made based on in vitro results, most of which tested at (extremely) high doses and excluded the use of proper positive controls or selectivity studies.

Future Research Recommendations

Regarding S. frutescens, the availability of in vivo data pertaining to its antiproliferative/anticancer activity appears to be lacking. In vivo studies are essential to determine the actual effectiveness of Sutherlandia and whether it can be safely and effectively incorporated into cancer treatment regimens. Only two documented clinical trials could be found, and it is important to note that they were to test tolerability [23] and were on HIV patients and not on cancer patients [12].

This literature review may be construed as overly critical or even negative, but in our view, the general poor quality of published research lends itself to the needed critique. Using established guidelines, current research indicates minimal to no in vitro activity of S. frutescens against the tested cancer cell lines. However, a common counterargument, driven by extensive anecdotal evidence and the herbʼs popularity, suggests potential in vivo activity, possibly due to active compounds being present at very low concentrations that standard bioassays will fail to detect. This hypothesis is plausible, given the well-documented chemical complexity of plant extracts, which can mask bioactivity through unwanted cellular interactions [55]. To conduct in-depth research on S. frutescence, techniques to reduce chemical complexity, as was published by the National Cancer Institute, which developed a rapid prefractionation procedure before in vitro bioactivity testing, should be employed [56]. Another explanation that is often provided for the lack of in vitro activity is that potential active compounds may exist as inactive prodrugs in the extract, undetectable in standard in vitro assays. Techniques to activate prodrugs for in vitro detection include incubation with human intestinal microbiota [57], enzymatic hydrolysis with β-glucosidase [58], incubation with liver microsomes [59], and various chemical hydrolysis techniques [60].

To guide future research on S. frutescens, the following stepwise approach is recommended:

1. Collect specimens from as many sources as possible to create sufficient chemical diversity.

2. Perform chemical fingerprinting (e.g., metabolomics) and in vitro bioactivity testing.

3. Prefractionate the most promising samples and repeat step 2.

4. If no significant bioactivity is detected, apply metabolic activation techniques and retest.

5. If these steps fail, consider in vivo animal studies or clinical trials, though ethical justification may be challenging, and hence, it must be concluded that S. frutescens lacks significant activity.

For well-characterised medicinal plants such as Cannabis sativa L. (Cannabaceae) containing the active cannabinoids, Artemisia annua L. (Asteraceae) containing artemisinin, or Mesembryanthemum tortuosum L. (Aizoaceae) containing mesembrine alkaloids, the known active compounds enable quality control and production of standardised preparations. In contrast, the unknown active compounds in S. frutescens preclude effective quality control, making the identification of active compounds, if present, a critical priority.

Moreover, the gap between in vitro and in vivo studies during medicinal plant drug discovery, due to the limitations of traditional in vitro models in replicating complex physiological conditions, hinders the progress of developing new therapies. Advanced in vitro models, such as organoids, organ-on-a-chip systems, and 3D bioprinted tissues, can accurately mimic human tissues, enabling precise analysis of plant compounds for therapeutic efficacy and potential toxicity [61], [62], [63]. Computational tools, including PK/PD modelling and even artificial intelligence platforms, can predict in vivo behaviour from in vitro data through analysis of complex datasets, correlating in vitro target inhibition with in vivo cell activity, which is particularly useful for identifying and evaluating plant-derived compounds. These models may mediate the translatability of in vitro results to in vivo outcomes by identifying patterns and predicting possible drug-target interactions [64], [65]. Ex vivo tissue models validate in vitro results, assessing compound absorption and metabolism [66]. Microphysiological systems and biomarker-driven assays enhance physiological relevance, while integrated platforms combining traditional knowledge and omics data streamline drug discovery [67], [68], [69]. Nanotechnology can address challenges like low solubility, thus improving translatability [64]. A phased approach, such as adopting advanced models, scaling multi-organ systems, and standardising platforms, could potentially enhance scientific research into S. frutescens and subsequently may lead to the development of plant-derived therapies.

Conclusions

Based on the available literature, we must conclude that research conducted to date on the antiproliferative effects of Sutherlandia is mainly promotional, with a few exceptions, such as that published by [27]. The level of obfuscation that was found in many publications is a serious cause of concern. Considering the serious nature of cancer, we find this to be unconscionable and alarming, and this negatively impacts this research field. Conclusions such as “partially validated” and “…anticancer activity of the S. frutescens … may even surpass that of the model chemotherapeutic drug, paclitaxel” are misleading, to say the least, considering the major limitations and shortcomings of almost all published studies. A simple rule of thumb is not to make any clinical claims based on in vitro data alone. A striking illustration of our ‘promotional research’ statement is how a negative report was reframed positively by referencing an early mention of Sutherlandiaʼs use nearly 100 years later. In the original text, as published in 1895, and somewhat typical of the time, it is stated that “There can be no reasonable doubt that Sutherlandia has been the means of curing malignant tumours, cancerous in appearance, which were firmly believed to be cancers by non-professional persons unacquainted with the distinctive marks of typical cancer. It is also certain, that employed as blood purifier and tonic, it has delayed the progress of true cancer and much prolonged life. In view of these facts and of the confident assertions made regarding Sutherlandia, it was desirable that it should be tested in a typical case of cancer, and that was done recently by a physician of high intelligence with the result, as in some similar instances formerly, that it proved wanting though applied both externally and internally” (boldface and underlining added) [5]. This paragraph was ‘cited verbatim’ from the original text but was presented as “curing of malignant tumours, cancerous in appearance; also used as blood purifier and tonic … to delay the progress of true cancer and much prolonged life” [4]. That it was tested and found wanting was simply omitted, and this is, in our view, promotional research.

Based on the available information and keeping in mind possible herb-drug interactions [70] and the serious nature of cancer, which causes over 10 million deaths annually, S. frutescens should currently not be recommended for use. To answer the question being asked in the title: currently, there seems to be no scientific merit in the anticancer claims made for S. frutescens, and research to date is mainly promotional.

Contributors’ Statement

CRediT authorship contribution statement Nicezelle Gernandt: writing – original draft, review and editing. Frank Van der Kooy: conceptualisation, supervision, data curation, writing – original draft, review and editing.

• References

• 1 Aboyade OM, Styger G, Gibson D, Hughes G. Sutherlandia frutescens: The meeting of science and traditional knowledge. J Altern Complement Med 2014; 20: 71-76
• 2 Chen L, Xu M, Gong Z, Zonyane S, Xu S, Makunga NP. Comparative cardio and developmental toxicity induced by the popular medicinal extract of Sutherlandia frutescens (L.) R. Br. detected using a zebrafish Tuebingen embryo model. BMC Complement Altern Med 2018; 18: 1-11
• 3 Viljoen A, Chen W, Mulaudzi N, Kamatou G, Sandasi M. Phytochemical Profiling of Commercial Important South African Plants. London: Academic Press, Elsevier; 2022
• 4 Van Wyk BE, Albrecht C. A review of the taxonomy, ethnobotany, chemistry and pharmacology of Sutherlandia frutescens (Fabaceae). J Ethnopharmacol 2008; 119: 620-629
• 5 Smith A. A Contribution to South African Materia Medica: Chiefly from Plants in Use Among the Natives. 2nd ed, Lovedale, South Africa; JC Juta & Co: 1895.
• 6 Watt JM, Breyer-Brandwijk MG. The Medicinal and Poisonous Plants of Southern and Eastern Africa. 2nd Edition. Edinburgh: E. and S. Livingstone Ltd.; 1962
• 7 Wicht WF. South African huismiddels. S Afr Med Rec 1918; 16: 306-310
• 8 Palmer E. The South African herbal. Cape Town: Tafelberg Publishers; 1985
• 9 Rood B. Uit die veldapteek. Cape Town: Tafelberg Publishers; 1994
• 10 Van Wyk BE, Gericke N. (eds.) Peopleʼs Plants. Pretoria, South Africa: Briza Publishers; 2000
• 11 Mills E, Cooper C, Seely D, Kanfer I. African herbal medicines in the treatment of HIV: Hypoxis and Sutherlandia. An overview of evidence and pharmacology. Nutr J 2005; 4: 1-6
• 12 Wilson D, Goggin K, Williams K, Gerkovich MM, Gqaleni N, Syce J, Bartman P, Johnson Q, Folk WR. Consumption of Sutherlandia frutescens by HIV-seropositive South African adults: An adaptive double-blind randomized placebo controlled trial. PLoS One 2015; 10: e0128522
• 13 Fu X, Li XC, Wang YH, Avula B, Smillie TJ, Mabusela W, Syce J, Johnson Q, Folk W, Khan IA. Flavonol glycosides from the South African medicinal plant Sutherlandia frutescens . Planta Med 2010; 76: 178-181
• 14 Fu X, Li XC, Smillie TJ, Carvalho P, Mabusela W, Syce J, Johnson Q, Folk W, Avery MA, Khan IA. Cycloartane glycosides from Sutherlandia frutescens . J Nat Prod 2008; 71: 1749-1753
• 15 Mavimbela T, Vermaak I, Chen W, Viljoen A. Rapid quality control of Sutherlandia frutescens leaf material through the quantification of SU1 using vibrational spectroscopy in conjunction with chemometric data analysis. Phytochem Lett 2018; 25: 184-190
• 16 Avula B, Wang YH, Smillie TJ, Fu X, Li XC, Mabusela W, Syce J, Johnson Q, Folk W, Khan IA. Quantitative determination of flavonoids and cycloartanol glycosides from aerial parts of Sutherlandia frutescens (L.) R. BR. by using LC-UV/ELSD methods and confirmation by using LC–MS method. J Pharm Biomed Anal 2010; 52: 173-180
• 17 Albrecht CF, Stander MA, Grobbelaar MC, Colling J, Kossmann J, Hills PN, Makunga N. LC–MS-based metabolomics assists with quality assessment and traceability of wild and cultivated plants of Sutherlandia frutescens (Fabaceae). S Afr J Bot 2012; 82: 33-45
• 18 Zonyane S, Chen L, Xu MJ, Gong ZN, Xu S, Makunga NP. Geographic-based metabolomic variation and toxicity analysis of Sutherlandia frutescens LR Br.–An emerging medicinal crop in South Africa. Ind Crops Prod 2019; 133: 414-423
• 19 Tchegnitegni BT, Lerata MS, Beukes DR, Antunes EM. Sutherlandiosides E−K: Further cycloartane glycosides from Sutherlandia frutescens . Phytochem Lett 2024; 61: 66-74
• 20 Dykman EJ. De Suid Afrikaanse Kook-, Koek- en Resepte Boek, 14th improved impression. Paarl (Cape Colony), South Africa: Paarl Printers Ltd.; 1908
• 21 Ojewole JAO. Anticonvulsant property of Sutherlandia frutescens R.Br. (variety incana E.Mey.) [Fabaceae] shoot aqueous extract. Brain Res Bull 2008; 75: 126-132
• 22 Seier J, Mdhluli M, Dhansay M, Loza J, Laubscher R. A toxicity study of Sutherlandia leaf powder (Sutherlandia microphylla) consumption. Parow South Africa: Medical Research Council (MRC); 2002
• 23 Johnson Q, Syce J, Nell H, Rudeen K, Folk WR. A randomized, double-blind, placebo-controlled trial of Lessertia frutescens in healthy adults. PLoS Clin Trials 2007; 2: e16
• 24 Sia C. Spotlight on ethnomedicine: usability of Sutherlandia frutescens in the treatment of diabetes. Rev Diabet Stud 2004; 1: 145
• 25 Tai J, Cheung S, Chan E, Hasman D. In vitro culture studies of Sutherlandia frutescens on human tumor cell lines. J Ethnopharmacol 2004; 93: 9-19
• 26 Chinkwo KA. Sutherlandia frutescens extracts can induce apoptosis in cultured carcinoma cells. J Ethnopharmacol 2005; 98: 163-170
• 27 Steenkamp V, Gouws MC. Cytotoxicity of six South African medicinal plant extracts used in the treatment of cancer. S Afr J Bot 2006; 72: 630-633
• 28 Stander BA, Marais S, Steynberg TJ, Theron D, Joubert F, Albrecht C, Joubert AM. Influence of Sutherlandia frutescens extracts on cell numbers, morphology and gene expression in MCF-7 cells. J Ethnopharmacol 2007; 112: 312-318
• 29 Stander A, Marais S, Stivaktas V, Vorster C, Albrecht C, Lottering ML, Joubert AM. In vitro effects of Sutherlandia frutescens water extracts on cell numbers, morphology, cell cycle progression and cell death in a tumorigenic and a non-tumorigenic epithelial breast cell line. J Ethnopharmacol 2009; 124: 45-60
• 30 Vorster C, Stander A, Joubert A. Differential signaling involved in Sutherlandia frutescens-induced cell death in MCF-7 and MCF-12A cells. J Ethnopharmacol 2012; 140: 123-130
• 31 Mqoco TV, Visagie MH, Albrecht C, Joubert AM. Differential cellular interaction of Sutherlandia frutescens extracts on tumorigenic and non-tumorigenic breast cells. S Afr J Bot 2014; 90: 59-67
• 32 Leisching G, Loos B, Nell T, Engelbrecht AM. Sutherlandia frutescens treatment induces apoptosis and modulates the PI3-kinase pathway in colon cancer cells. S Afr J Bot 2015; 100: 20-26
• 33 Lin H, Jackson GA, Lu Y, Drenkhahn SK, Brownstein KJ, Starkey NJ, Lamberson WR, Fritsche KL, Mossine VV, Besch-Williford CL, Folk WR. Inhibition of Gli/hedgehog signaling in prostate cancer cells by “cancer bush” Sutherlandia frutescens extract. Cell Biol Int 2016; 40: 131-142
• 34 van der Walt NB, Zakeri Z, Cronjé MJ. The induction of apoptosis in A375 malignant melanoma cells by Sutherlandia frutescens . Evid Based Complement Alternat Med 2016; 2016: 4921067
• 35 Omoruyi SI, Enogieru AB, Ekpo OE. Cytotoxic and apoptosis-inducing effects of Sutherlandia frutescens in neuroblastoma cells. J Afr Assoc Physiol Sci 2018; 6: 136-144
• 36 Zonyane S, Fawole OA, La Grange C, Stander MA, Opara UL, Makunga NP. The implication of chemotypic variation on the anti-oxidant and anti-cancer activities of Sutherlandia frutescens (L.) R.Br. (Fabaceae) from different geographic locations. Antioxidants 2020; 9: 152
• 37 Smit T. Anticancer efficacy of selected South African phytomedicines in a three-dimensional colorectal cancer model [Dissertation]. Potchefstroom: North-West University; 2019
• 38 Gouws C, Smit T, Willers C, Svitina H, Calitz C, Wrzesinski K. Anticancer potential of Sutherlandia frutescens and Xysmalobium undulatum in LS180 colorectal cancer mini-tumors. Molecules 2021; 26: 605
• 39 Motadi LR. Screening methanolic extracts of Sutherlandia spp (Cancer Bush) as anti-tumor agents and their effects on anti-apoptotic genes: Anti-tumor of Sutherlandia spp. Iran J Pharm Res 2020; 16: 45-58
• 40 Van der Kooy F. Artemisia afra and COVID-19, the media storm versus the current state-of-the-art. Rev Bras Farmacogn 2024; 34: 1265-1275
• 41 Verpoorte R, Houghton PJ, Heinrich M, Mukherjee PK, Schmeda Hirschmann G, van Staden J, Yesilada E. Editorial. J Ethnopharmacol 2006; 103: 309-310
• 42 Petersen EJ, Nguyen A, Brown J, Elliott JT, Clippinger AJ, Gordon J, Kleinstreuer N, Roesslein M. Characteristics to consider when selecting a positive control material for an in vitro assay. ALTEX 2021; 38: 365-376
• 43 Skerman NB. The apoptosis inducing effects of Sutherlandia spp. extracts on an oesophageal cancer cell line. [Dissertation]. Johannesburg: University of Johannesburg; 2011
• 44 Butterweck V, Nahrstedt A. What is the best strategy for preclinical testing of botanicals? A critical perspective. Planta Med 2012; 78: 747-754
• 45 Vogel D, Loots E, Oladimeji O, Gouws C, van der Kooy F. The anti-neoplastic activity of Artemisia afra in breast cancer cell lines. S Afr J Bot 2023; 157: 115-121
• 46 Agarwal A, DʼSouza P, Johnson TS, Dethe SM, Chandrasekaran CV. Use of in vitro bioassays for assessing botanicals. Curr Opin Biotechnol 2014; 25: 39-44
• 47 Gertsch J. How scientific is the science in ethnopharmacology? Historical perspectives and epistemological problems. J Ethnopharmacol 2009; 122: 177-183
• 48 Pritchetta JC, Naesens L, Montoya J. Treating HHV-6 infections, the laboratory efficacy and clinical use of anti-HHV-6 agents. In: Flamand L, Lautenschlager I, Krueger G, Ablashi D (eds.) Human Herpes Viruses HHV-6A, HHV-6B, and HHV-7. Diagnosis and Clinical Management. 3th ed. Amsterdam: Elsevier; 2014: 311-331
• 49 Indrayanto G, Putra GS, Suhud F. Validation of in-vitro bioassay methods: Application in herbal drug research. Profiles Drug Subst Excip Relat Methodol 2021; 46: 273-307
• 50 López-Lázaro M. Two preclinical tests to evaluate anticancer activity and to help validate drug candidates for clinical trials. Oncoscience 2015; 2: 91
• 51 López-Lázaro M. How many times should we screen a chemical library to discover an anticancer drug?. Drug Discov Today 2015; 2: 167-169
• 52 Cos P, Vlietinck AJ, Berghe DV, Maes L. Anti-infective potential of natural products: How to develop a stronger in vitro ‘proof-of-concept’. J Ethnopharmacol 2006; 106: 290-302
• 53 Katt ME, Placone AL, Wong AD, Xu ZS, Searson PC. In vitro tumor models: Advantages, disadvantages, variables, and selecting the right platform. Front Bioeng Biotechnol 2016; 12: 4-12
• 54 Zang R, Zhang X, Sun J, Yang ST. In vitro 3-D multicellular models for cytotoxicity assay and drug screening. Process Biochem 2016; 51: 772-780
• 55 van der Kooy F, Sullivan SE. The complexity of medicinal plants: The traditional Artemisia annua formulation, current status and future perspectives. J Ethnopharmacol 2013; 150: 1-13
• 56 Grkovic T, Akee RK, Thornburg CC, Trinh SK, Britt JR, Harris MJ, Evans JR, Kang U, Ensel S, Henrich CJ, Gustafson KR, Schneider JP, OʼKeefe BR. National Cancer Institute (NCI) program for natural products discovery: Rapid isolation and identification of biologically active natural products from the NCI prefractionated library. ACS Chem Biol 2020; 15: 1104-1114
• 57 Atanasov AG, Waltenberger B, Pferschy-Wenzig EM, Linder T, Wawrosch C, Uhrin P, Temml V, Wang L, Schwaiger S, Heiss EH, Rollinger JM, Schuster D, Breuss JM, Bochkov V, Mihovilovic MD, Kopp B, Bauer R, Dirsch VM, Stuppner H. Discovery and resupply of pharmacologically active plant-derived natural products: A review. Biotechnol Adv 2015; 33: 1582-1614
• 58 Tahtah Y, Wubshet SG, Kongstad KT, Heskes AM, Pateraki I, Møller BL, Jäger AK, Staerk D. High-resolution PTP1B inhibition profiling combined with high-performance liquid chromatography–high-resolution mass spectrometry–solid-phase extraction–nuclear magnetic resonance spectroscopy: Proof-of-concept and antidiabetic constituents in crude extract of Eremophila lucida . Fitoterapia 2016; 110: 52-58
• 59 Tu Y. The discovery of artemisinin (Qinghaosu) and its derivatives: From traditional Chinese medicine to modern antimalarial drugs. Chin Med J 2016; 129: 2429-2435
• 60 Cui S, Wang X, Wang G, Wang R, Yang C. A network pharmacology approach to investigate the anti-inflammatory mechanism of effective ingredients from Salvia miltiorrhiza . Int Immunopharmacol 2020; 81: 106040
• 61 Wang L, Hu D, Xu J, Hu J, Wang Y. Complex in vitro model: A transformative model in drug development and precision medicine. Clin Transl Sci 2024; 17: e13695
• 62 Kostrzewski T, Maraver P, Ouro-Gnao L, Levi A, Snow S, Miedzik A, Rombouts K, Hughes D. A microphysiological system for studying nonalcoholic steatohepatitis. Hepatol Commun 2020; 4: 77-91
• 63 Ma X, Liu J, Zhu W, Tang M, Lawrence N, Yu C, Gou M, Chen S. 3D bioprinting of functional tissue models for personalized drug screening and in vitro disease modeling. Adv Drug Deliv Rev 2018; 132: 235-251
• 64 Hu F, Wang D, Huang H, Hu Y, Yin P. Bridging the gap between target-based and cell-based drug discovery with a graph generative multitask model. J Chem Inf Model 2022; 62: 6046-6056
• 65 Najmi A, Javed SA, Al Bratty M, Alhazmi HA. Modern approaches in the discovery and development of plant-based natural products and their analogues as potential therapeutic agents. Molecules 2022; 27: 349
• 66 Xu Y, Shrestha N, Préat V, Beloqui A. An overview of in vitro, ex vivo and in vivo models for studying the transport of drugs across intestinal barriers. Adv Drug Deliv Rev 2021; 175: 113795
• 67 Sengupta A, Schmid S, Grangier N, Dorn A, Hebestreit M, Hugi A, Žajdlíková K, Herbst A, Losada-Oliva P, Ortolf-Wahl H, Krebs P. A next-generation system for smoke inhalation integrated with a breathing lung-on-chip to model human lung responses to cigarette exposure. Sci Rep 2025; 15: 1-9
• 68 Katiyar C, Gupta A, Kanjilal S, Katiyar S. Drug discovery from plant sources: An integrated approach. Ayu 2012; 33: 10-19
• 69 Chen T, Yang M, Cui G, Tang J, Shen Y, Liu J, Yuan Y, Guo J, Huang L. IMP: Bridging the gap for medicinal plant genomics. Nucleic Acids Res 2024; 52: D1347-D1354
• 70 Fasinu PS, Gutmann H, Schiller H, James AD, Bouic PJ, Rosenkranz B. The potential of Sutherlandia frutescens for herb-drug interaction. Drug Metab Dispos 2013; 41: 488-497