Coley’s Toxin

The idea of inducing an infection to promote collateral regression of tumours gained momentum with the advent of Coley’s Toxin. In 1891, Dr. William Bradley Coley, an American bone surgeon and cancer researcher, developed a formulation, the Coley’s Toxin. It consisted of heat-inactivated Streptococcus pyogenes and Serratia marcescens. At New York’s Memorial Hospital, Coley used the vaccine to treat a sarcoma patient successfully [10]. After that, many cancer patients underwent treatment with the Toxin with promising results. Even when administered at the advanced stages of cancer, the Toxin brought forth remarkable recoveries [10,11]. Nevertheless, for the formulation to be effective, Coley maintained, a few conditions were to be met, including induction of an acute infection that makes the patient feverish with considerably elevated body temperature [12]. The inoculation should be given daily or on every alternate day for a month or two with gradually increasing dosage. The Toxin should be injected directly into the tumour. Further, a six-month course of weekly injections is needed to avoid relapse [12]. Coley’s Toxin was effective against many cases of sarcoma, lymphoma, melanomas, and myelomas. Although it was a breakthrough treatment strategy, with time, advanced treatment modalities, such as radiation therapy and chemotherapy, gained credence and momentum. A lack of understanding of the exact mechanism of action of the Toxin, the risks associated with infecting patients with potentially pathogenic bacteria, and varied and often unpredictable responses to the treatment led eventually to the relinquishment of this treatment strategy in its original form. The infection-induced remission of tumours nevertheless shed light on the vital roles of the immune system in fighting cancer, as substantiated by prominent scientists of that period, including Paul Ehrlich [9]. The last successful use of Coley’s Toxin was reported in China in 1980 when a patient with terminal liver cancer was given 68 injections of the Toxin for 34 weeks [11,13]. Coley’s efforts have laid the foundation for developing anticancer immunotherapy regimens. He is deservedly considered as father of cancer immunotherapy.

BCG (Bacillus Calmette-Guerin) and SR

During the 20^th century, substantial progress in the field of immunology enabled researchers to revisit Coley’s treatment strategy with a new perspective and deepened insight. Inspired by the findings of Coley, for instance, Lloyd Old and colleagues repurposed the well-known antituberculosis bacterial formulation, Bacillus Calmette-Guerin (BCG). It comprises an attenuated strain of Mycobacterium bovis [14] (BCG was developed by two French Scientists, Albert Calmette and Camille Guerin, in 1921). During the 1950s, Old studied the effect of BCG on solid tumours and Ehrlich ascites cells implanted in mice [15]. The formulation regressed the tumours considerably.

Further evidence of BCG’s efficacy came in 1976 when many recurrent superficial bladder tumours were treated successfully with it [16]. A timeline of these and subsequent developments is available in Fig. 1. Later studies attested the utility of BCG to induce tumour remission via activating the immune system. Although the precise anti-tumour mechanism of BCG is not clear, the internalization of BCG by the tumour cells, the subsequent display of antigens on their surface and the consequent release of cytokines, such as IL-2, TNF-α, INF-γ, IL-4, and IL-6 from infiltrating T-lymphocytes, are thought to induce the remission (Fig. 2). Supporting this postulate, mice depleted of CD4+ and CD8+ T-cells exhibited only a negligible response to BCG [17]. In addition to the T-cells, granulocytes, macrophages, and dendritic cells also play pivotal roles in the remission; the number of these cells increases considerably during the treatment [18], [19], [20]. BCG therapy, nevertheless, has its caveats. After the remission, the tumour returned in nearly 30% of the patients. Nevertheless, by far, BCG is the only bacterial therapeutic used for the treatment of non-muscle invasive bladder cancer (NMIBC) for the past 40 years [18].

1. Download : Download high-res image (568KB)
2. Download : Download full-size image

Fig. 2. Postulated anti-tumour mechanism of BCG in non-muscle invasive bladder cancer. BCG gets internalized in the cancer cells of the urothelium and and in the antigen-presenting cells (APCs). The urothelial cells and the activated APCs present the BCG antigens on their surface, leading to cytokine/chemokine release and granuloma formation. These granulomas, comprising macrophages, dendritic cells (DCs), neutrophils, T cells, B cells, and fibroblasts, trigger innate and adaptive immune responses and destroy the cancer cells.

Oncolytic viruses (OV)

One of the earliest examples of virus-orchestrated SR was reported in 1896 when influenza cured a patient of leukaemia (Fig. 1) [21]. Measles-induced remission of Burkitt’s and Hodgkin lymphoma [22,23] and several similar cases were reported subsequently [24]. SR after viral infections thus paved the way for the development of a new mode of cancer treatment- cancer virotherapy. Initially (between 1950–60), wild-type viruses were used to induce the remission, which worked successfully in some cases but proven to be dangerous to many patients due to uncontrolled infection [25]. The risks associated with the use of natural viruses led eventually to the dissolution of this treatment modality. However, by the end of the 20th century, the field of cancer virotherapy regained propulsion with the advent of genetically modified viruses [9]. The inherent ability of viruses to enter specific host cells (cellular tropism), to replicate therein and to subsequently lyse the host, and to exert bystander toxicity to the neighboring cells make them potential therapeutic agents (Fig. 3). Furthermore, upon exposure of the viral PAMPs (Pathogen-associated molecular patterns), the immune system primes the T-cells to effect contemporaneous elimination of metastasized tumours (the so-called “abscopal effect”) [26] (Fig. 3). Viral species such as adeno, herpes, vaccinia, polio, and measles are design-optimized and evaluated in clinical settings as monotherapy or as combination therapies [27]. Talimogene laherparevec (T-VEC) is the first engineered oncolytic herpes simplex virus (HSV) that received United States Food and Drug Administration’s approval for the treatment of advanced melanoma [28,29]. T-VEC is a double-edged weapon that not only destroys the target tumour cells at the site of injection but could also initiate localized as well as systemic immune response [30]. Locally, the lysed tumour cells release tumour-associated antigens. These antigens attract members of the mononuclear phagocyte system-especially the macrophages-promoting clearance of the dying and dead tumour cells. The dendritic cells then present these viral antigens to T-helper cells and cytotoxic T-cells and thereby mount a formidable immune response at the systemic level [30]. Insights into the mechanism of action of T-VEC have also led to the development of novel combinatorial therapeutics. For example, T-VEC, in combination with ipilimumab (an inhibitor of the immune-suppressor protein, CTLA-4), has entered clinical trials for melanoma (https://clinicaltrials.gov/ct2/show/NCT01740297). Similarly, the efficacy of T-VEC in combination with pembrolizumab (a monoclonal antibody that blocks another immune checkpoint protein PD-1) to improve the outcome of melanoma is currently under investigation (https://clinicaltrials.gov/ct2/show/NCT02263508).

1. Download : Download high-res image (692KB)
2. Download : Download full-size image

Fig. 3. Oncolytic viruses. Many OVs selectively replicate in cancer cells (cellular tropism) and lyse the cells. The resultant release of pathogen-associated molecular patterns (PAMPs) and damage-associated molecular patterns (DAMPs) recruits and activates dendritic cells (DCs). DCs then migrate to lymph nodes and present the antigens to the T cells. These activated cytotoxic T cells undergo clonal expansion. They induce apoptosis in target cells via the release of perforin and granzymes.

SR of specific cancers

Leukaemia, which accounts for 3.1% of global cancer deaths [1], is another cancer that is prone to SR. Bradley and colleagues reviewed several cases of SR of leukaemia that were reported from 1949 to 2019 [34]. Fifty cases were of acute myeloid leukaemia (AML), and four were acute lymphocytic leukaemia (ALL). For leukaemia, out of the five cases reported in 2018, infection-driven remissions were seen in four cases (Table 1). Unfortunately, many of the remissions of AML were not long-lasting: a high relapse rate (< 1Y) was reported among the patients [34,35]. Nevertheless, design-optimized, infection-based therapies for AML might yield a better prognosis. During the last leg of eradication of smallpox, it has been observed that the smallpox vaccine could induce remission of chronic lymphocytic leukaemia [36]. In another instance, an acute flare-up of hepatitis B virus healed a patient of splenic marginal-zone lymphoma [37]. The virus build-up and consequent immune response cleared atypical lymphocytes from the blood and bone marrow of the patient. Concerning SR of lymphoma, three cases of diffuse large B-cell lymphoma (DLBCL), one of follicular lymphoma (FL), and one of Hodgkin lymphoma (HL) (Tables 1, 2) were reported in the past three years. Some of these SRs-especially those associated with FL-could be attributed to Helicobacter pylori infection, a known activating factor in the patient’s immune response against the lymphoma cell [38]. DLBCL, on the other hand, is an aggressive cancer with a less prognostic probability of SR. Nevertheless, such remissions do occur. For example, Tanaka and colleagues presented a case of SR of DLBCL [39]. In the patient’s biopsy sample, rampant apoptosis of the cancer cells were observed, suggesting that the regression could have occurred via T-cell-mediated induction of apoptosis [39].

Lung cancer accounts for 18% of cancer deaths, making it the leading cause of cancer death [1]. Zhang et al. reported 14 cases of remission of lung cancer, barring patients who have received any chemotherapy and radiotherapy [32]. Among cancers of the lung, SR of non-small cell lung cancer (NSCLC) is very rare. Interestingly, among the reported SR cases of lung cancer, none showed an association with concurrent or preceding infections (Table 2). Instead, such regressions were associated with postoperative trauma. Further, one of the patients attested using the herbal extract of Rock pine (Orostachys japonicus) [40] as a potential causative agent for the remission. Rock pine, widely grown on the rocks in Korea, Japan, and China, has several anticancer compounds [41]. The possibility that some of these compounds in the extract contributing to the remission cannot be ruled out.

Merkel cell carcinoma (MCC) is an aggressive skin cancer with a frequency of 0.7 out of one lakh population and shows a high rate of metastasis [42]. Several cases of SR have been documented for MCC in the past three years. Other less common cases of SR include the regression of cancers of the liver, kidney, and colorectal areas (Table 2).

Decoding SR with a therapeutic perspective

Different forms of cancer originate via a complex interplay among signaling cascades induced by genomic, metabolomic, and proteomic aberrations. Further, their often-counterintuitive modes of sustenance and progression mean that the spontaneous remission of these cancers could as well involve diverse inputs and mechanisms (Fig. 4). Febrile infections are responsible for immune cell infiltrations and cytokine cascades, leading to SR in many cases. Nevertheless, as we have mentioned before, not all cases of SR can be traced to bacterial or viral infection (Table 2). In such scenarios, there must exist hitherto poorly understood mechanisms that heal the tumour. Based on medical and scientific observations, some possible whys and wherefores are as follows: Partial excision of a tumour impairs blood supply to the remaining malignant tissue. Since tumour cells require an ample supply of blood (so much so that the cells themselves promote angiogenesis to meet this vampiric need), cutting it off could starve the cells to death [4]. Not surprisingly, therefore, many SR cases have been attributed to biopsy procedures [4]. Another interesting possibility lies in tumour ablation procedures that could set forth an avalanche of tumour-derived antigens in circulation. Studies have shown that the antigens thus released could act as a therapeutic vaccine [64] (Fig. 4).

1. Download : Download high-res image (550KB)
2. Download : Download full-size image

Fig. 4. Possible mechanisms involved in SR of cancers. Factors that stimulate an immune response, including fever and infection, may promote SR. Similarly, disruption in the tumour microenvironment, biopsy and ablation procedures and the induction of cell death by various means, may also induce SR.

Techniques that use radiofrequency waves, high-intensity focused ultrasound, or cryoablation are potential activators of immune cells [64]. In addition to its hormetic effect, exposure to small doses of radiation (such as those received during radiographic examinations) has been shown to induce tumour remission [65]. Low-level radiation (≤ 0.1 Gy for acute irradiation, for example) is capable of pro-SR immunomodulation in experimental settings. Whole- or half-body exposures to low dose radiations [66] as enablers of anticancer immune defense is gaining considerable momentum. Blood transfusion procedures performed in leukemic patients, and use of herbal extracts are also implicated as potential triggers of SR [67,40]. Hormonal influence in SR of breast and prostate cancers and withdrawal of carcinogenic agents could promote the remission [4]. Such non-targeted bystander inputs could lead to processes such as telomerase inhibition, promoting cytostatic or cytocidal outcomes in cancer cells [24] (Fig. 4). Concerning the latter, not only apoptosis but non-apoptotic cell death pathways [68] could also facilitate SR. Such pathways can be considered for future interventional or surrogate therapeutic strategies. Changes in the tumour microenvironment (TME) due to presence of metalloproteinase inhibitors and anti-angiogenesis factors are also believed to play a role in the regression [69] (Fig. 4).

Kwok et al. analyzed SR cases of chronic lymphocytic leukaemia (CLL). They excluded patients with infections, second malignancies, or are undergoing immunosuppressive therapies from their study [70] and compared regressing and non-regressing CLL using flow-cytometry, single nucleotide polymorphism (SNP) analysis, and exome sequencing. It was found that CLL clone in residual SR tumours gradually acquires a low-proliferation state with reduced expression of migration-specific markers. Therefore, it is postulated that in specific genomic backgrounds, changes in microenvironmental factors, such as B-cell-receptor stimulation, can spontaneously regress CLL [70]. This is the only major study published to date that had analyzed the transcriptomic and genomic data to understand the intricacies of SR.

SR and immunity

Since immune response plays crucial roles in tumour regression, many SR cases could also be considered as immunity-mediated tumour remissions. In fact, SR linked to infections influenced the discovery of different anti-cancer therapies that work by boosting the immune system to recognize and attack the cancer cells [71]. The field of cancer immunotherapy (CI) has evolved from clinically-induced infections (Coley’s toxins) in the 19th century and cytokine therapy and cancer vaccines in the 20th century to current approaches, such as immune checkpoint inhibitors, oncolytic viruses and Chimeric antigen receptor (CAR)-T cell therapy [72,9]. Different CIs-although they differ in their mechanism of action-work to bolster a weakened immune system. Immuno-onco therapies are broadly classified into active and passive therapies [72]. Active immunotherapy involves triggering the host immune response using cytokines or cancer-specific antigens leading to immunological memory. Passive therapies include direct targeting of tumours with monoclonal antibodies (mABs) or modified immune cells, as in the case of adoptive cell transfer therapy (ACT), without inducing immunological memory [73]. Establishing CI as the fourth pillar of cancer treatment modality (along with chemotherapy, radiation and surgery), the number of CI drugs has been on the rise and many molecules are undergoing clinical trials [74]. CI nevertheless has its drawbacks, such as immunological toxicity [75]. Studies showing that gut microbiomes could influence an individual’s anti-tumour response to immunotherapeutic agents are quite intriguing [75], [76], [77]. It has been reported that bacteria from the Bifidobacteriaceae and Bacteroidaceae family enhance the response to anti-PD-L1 (anti-programmed cell death protein ligand- 1) and anti-CTL4 (anti-cytotoxic T-lymphocyte antigen 4S) therapies, respectively [78,79].

Future perspective

Occurrences of spontaneous regression have been reported for many types of cancer. Despairingly, the frequency of SR is very low; one in 60,000–1,00,000 cancer cases [4,9,80]. Melanomas, lymphoma, leukaemia, neuroblastoma, renal cell carcinoma, and germ cell cancers have shown higher frequency of SR [81,8]. The reasons behind higher percentages of SR in certain types of cancer are poorly understood. Understanding the mechanistic aspects of SR of cancer could pave the way for innovative strategies to improve cancer outcomes.

Learning from the drawbacks of Coley’s toxin, the search for efficacious recombinant microbes as cancer therapeutic agents began in the 21st century [82]. The genetically engineered bacteria thus created express cytotoxic proteins or toxins to elicit an anti-tumour immune response. For example, programmable bacteria could grow into the core of tumours and deliver anti-CD47 nanobodies into B-cell-lymphoma [83]. A welcome addition to this targeted repertoire is a class of engineered non-pathogenic bacteria that can prime cytotoxic T-cells to induce tumour regression [84]. Anaerobic bacteria could be an apt choice to enhance the targeted delivery into the deep interiors of solid tumours due to their penchant for hypoxic environments [85,82].

Given that antibiotics are available for most bacterial strains, it might be easier to regulate the possibility or severity of infection in such treatment regimens. Interestingly, in addition to full-grown bacteria, bacterial spores have been explored for their tumoricidal effect. When the spores of certain bacterial species germinate, get metabolically active, and grow under conducive conditions, they could promote tumour regression. In this context, Clostridium novyi NT (non-toxic) spores have been extensively studied. These spores have entered phase 1 clinical trials-alone or in combination with chemotherapeutic drugs-for the treatment of solid tumours that are not responding to conventional treatments [82]. When it comes to the fate of the tumour cells under self-regression, apoptosis could be the prevalent mechanism. However, exceptions do occur. During SR of some neuroblastoma, for example, the tumour cells undergo a RAS-associated, caspase-independent, non-apoptotic form of cell death [86]. Therefore, in addition to apoptosis, regulated cell death (RCD) mechanisms, such as ferroptosis, necroptosis, pyroptosis, parthanatos, etc. [68] may play a crucial role in SR. Given that many tumour cells resist natural or drug-induced apoptosis, exploring non-apoptotic RCD mechanisms involved during the self-regression would provide valuable insights of therapeutic relevance. As we have discussed earlier, understanding of infection-based immune system activation against tumours has led to the development of more efficacious treatment strategies of bacteriotherapy and virotherapy. However, research on the non-infection-based mechanisms of SR is still in its infancy.

It is quite challenging to deduce all the factors responsible for spontaneous or indirectly induced tumour regression. Bio-banking tumour biopsy samples of those patients who experienced SR without undergoing any cancer-specific treatments will allow retrospective studies [44]. Investigating SR case reports in the light of diet and microbiome composition of the recovered patient may aid in developing novel therapies with reduced side effects. Further, insights from these studies may help predispose and orient one’s lifestyle towards conditions that are barren for cancer cells to flourish.