Unfortunately, it is easier for us to detect the acute effects of a viral infection than to detect the long-term role that viral infection plays in our health. As a result, modern medicine has led us to make a number of decisions that are ultimately detrimental to our health.
The idea of febrile infections conferring protection against the development of tumors began to develop in the late 19th and early 20th century. Initially, anecdotal evidence based on physicians’ observations and examination of medical histories of patients suggested beneficial effects of infections on cancer risk.
More formalized data collection followed and in 1912 a study demonstrated that United States and Canadian Native American populations that had six times higher mortality rate from infectious disease had a lower rate of mortality from cancer (74).
In 1916, another study examined mortality rates in New York, Boston, Philadelphia, and New Orleans from 1864-1888 and 1889-1913. Over these two time periods, deaths from infectious disease decreased while simultaneously cancer death rates increased by over 55%.
In more contemporary times, a study examined the relationship between cancer and infectious disease rates in Italy (75). The authors proposed that there were four factors to consider when investigating possible causality in this relationship: 1. Consistency and strong associations among studies, 2. Temporality of the association, 3. True plausibility when considering the field of research, and 4. A dose response relationship between agent and disease.
According to them, temporality was the 12 most important aspect in determining association. With this as a guide, they attempted to resolve the conflict between evidence that infections cause cancer versus evidence that infections can prevent cancer.
Establishing a stricter temporal relationship between observations of a decrease in infections in Italy and observation of an increased cancer rate confirmed that in the first half of the 20th century the cancer increase indeed followed the decrease in infectious disease rates (75).
As evidence began to mount in favor of infections having a beneficial effect on preventing cancer, so did anecdotal evidence that concurrent febrile infection in cancer patients led to a better cancer control. Dr. William Coley was the first to administer infectious agents to a substantial number of cancer patients.
Coley administered a mixture of heat killed Streptococcus pyogenes and Serratia marrescens to late-stage sarcoma patients. This mixture, known as Coley’s Toxins, lead to cancer remissions in many sarcoma patients (76).
Even today, infection of the bladder with Mycobacterium bovis Bacillus Calmette-Guerin (BCG) is the treatment of choice for patients with non-muscle invasive bladder cancer (77). These bacterial infectious agents are thought to act as adjuvants enhancing ongoing anti-tumor immune responses.
While this may be the case, infections might play other roles in the development of anti-tumor immune responses. These novel roles of infections will be investigated later in this dissertation. In exploring relationships between childhood infections and cancer, the majority of studies focused on influenza, measles, mumps, pertussis/whooping cough, chicken pox, scarlet fever, rubella, and diphtheria.
A study done in 1966 was designed to examine the relationship between exposure to X-ray radiation and hormonal therapy and increasing mortality rates from ovarian cancer. 97 patients with ovarian cancer and 97 controls with benign ovarian tumors were recruited and a thorough medical history was obtained.
In the final analysis, X-ray radiation and hormone therapy were not found to have an impact on ovarian cancer mortality rates. Unexpectedly, 13 however, individuals who had reported a past history of mumps parotitis had a significantly reduced risk of ovarian cancer (p=0.0007)(78).
This study was the first to suggest that mumps infection could lead to lower incidence of ovarian cancer. The mechanism, however, remained unknown until Cramer et al proposed an immune mechanism in 2010 (62).
In 1977, another study addressed possible causes for the rising incidence of ovarian cancer. 300 women with laparotomy-confirmed ovarian cancer were recruited in 17 medical centers.
They were administered a questionnaire that included inquiries on X-ray exposure, pregnancies, smoking, chronic illness, history of acute infections, contraception use and prior surgeries, among others. The study included two control groups: 1. Gynecological patients in the hospital for reasons other than suspected ovarian cysts or tumors and 2. A group of women in the community provided from a list generated by general practitioners. These two control groups were provided with the same questionnaire as the case group.
When the data was compiled, many women diagnosed with ovarian cancer reported fewer past bouts of mumps, measles, chicken pox, and rubella. The study also displayed a statistically significantly reduced relative risk of developing ovarian cancer in the second control group among women who had a history of measles (Relative Risk (RR)=0.47), mumps (RR=0.61), rubella (RR=0.62), and chicken pox (RR=0.66) (79).
A recent retrospective review confirmed the outcome of reduced risk for ovarian cancer correlating with those four infections (74). Kömel et al. performed a case control study in 1992 in support of his hypothesis that childhood and/or adulthood febrile infections influenced a higher risk of melanoma.
The study was carried out at the University of Göttingen on 139 newly diagnosed melanoma patients whose primary tumor was removed between January 1988 and September 1991. The study also included 271 controls recruited from a pool of Ophthalmology and Dermatology clinic patients whose 14 diseases were not related to malignant melanoma.
Both cases and controls were age and gender matched. These groups were administered questionnaires to collect information on frequency and temperature of febrile infections experienced as well as questions on childhood, adulthood infections, and common infections within the 5 years of melanoma diagnosis.
Group one in the study consisted of individuals who had one or more of the following: measles, mumps, rubella, chicken pox, scarlet fever, diphtheria, whooping, and tonsillectomy performed prior to the age of 13 (assumed to have tonsillitis).
Group two were individuals who had infections in adulthood associated with fever that included hepatitis, tuberculosis, erysipelas, chronic infectious diseases, febrile abscess, furunculosis, wound infections, tropical diseases, and other fever producing diseases.
Common diseases accompanied by a fever within the five years of melanoma diagnosis (group three) included gastroenteritis, influenza/common cold, pneumonia, herpes simplex virus 1, and “other trivial diseases” associated with fever. Individuals in Group one were not found to have statistically significantly reduced risk of melanoma. However, Group two with the history of chronic infectious diseases (adjusted Odds Ratio (OR)=0.32) and febrile abscesses/furunculosis/wound infections (adjusted OR=0.21), were found to have a significantly reduced risk for melanoma.
Individuals in group three, particularly with a history of influenza/common cold (adjusted OR=0.32) and HSV1 (adjusted OR=0.45) also had significant risk reduction for melanoma. There was also evidence of the cumulative effects of multiple febrile infections compared to never experiencing a serious febrile infection. The study showed that more fevers experienced in group two (p=0.01) and group three (p=0.0001) translated to a significant risk reduction.
Although the calculated risk for melanoma was not significantly reduced in group one, childhood diseases such as chicken pox, measles and mumps had odds ratios <1 indicating a possible reduction of melanoma risk (80). 15 A repeat of this study on a larger number of cases included 603 melanoma patients from 11 medical centers linked via the European Organization for Research and Treatment of Cancer (EORTC) Melanoma Cooperative Group.
The control group included 627 individuals who were age matched and selected from the same neighborhoods of the melanoma cases. Individuals in this study were grouped according to a history of severe diseases (group one- hepatitis, tuberculosis, Staphylococcus aureus infections, urinary tract infections, sepsis, meningitis, rheumatic fever, cholecystitis, and erysipelas), and less severe infections suffered five years prior to primary tumor removal (group two- influenza, infectious enteritis, bronchitis, pneumonia, and HSV).
Results from this study confirmed the relationship between strong febrile infections accompanied by a body temperature above 38.5°C, and risk reduction for melanoma (81). Association between gliomas and Varicella-Zoster Virus (VZV) was examined in the San Francisco Bay Area Glioma Study. Individuals were questioned on their shingles and chicken pox histories.
These diseases caused by VZV were of particular interest for a link to glioma due to the virus having neurological sequela. The group found that glioma patients were less likely to have had a history of shingles or chicken pox compared to the control group. In order to have a more reliable proof of infection than self-reporting, a repeat of this study was performed on newly diagnosed glioma patients from whom the self-reported history of having chicken pox was obtained and confirmed serologically by positivity for anti-VZV antibodies.
Data obtained on 462 glioma subjects and 443 controls matched for age, sex, and ethnicity demonstrated a reduced risk for developing glioma in individuals with a history of VZV (OR=0.4). The importance of having a more reliable measure than self-reporting, such as serologic markers for this type of studies was justified by the difference in this study between self-reported chicken pox infections and 16 seropositivity for VZV. Individuals who were IgG seropositive for VZV had a reduced risk (OR=0.6) for having a glioma (82).
Albonico et al. studied a cohort of cancer patients with solid epithelial tumors diagnosed by 35 general practitioners in Switzerland, matched them according to age, gender, and physician with a recruitment limit of 20 patients/physicians office. A total of 410 patients were administered questionnaires that requested information on age, gender, history of febrile childhood infections (in this study defined as measles, mumps, scarlet fever, pertussis, rubella, and chicken pox), and frequency of other infections that resulted in fever >39°C prior to the age of 21.
Additionally, when answering questions specifically on febrile childhood infections; answering options were limited to “yes”, “uncertain yes,” “uncertain no,” and “no” in order to reduce recall/memory bias. Almost 50% were breast cancer patients and therefore the final analysis was carried out on breast cancers vs. non-breast cancers. Analysis was also divided according to age ≤60 vs. >60. Ultimately the study found a statistically significant reduction of risk of solid malignancy in individuals who had a history of chickenpox (p=0.044) or rubella (p=0.0003) (83).
The risk was further lowered with the increased number of infections. Curiously these results were obtained only in the 50% of individuals who had cancers other than breast cancer. Cancer risk reduction due to the history of febrile diseases did not hold for breast cancer patients, leading the authors to suggest that beneficial protection provided by childhood infections might be body site specific. Additional studies on patients with many different cancers came up with similar results (74, 83, 84).
One case-control study demonstrated that chicken pox and pertussis infections lowered risk for stomach, breast, colorectal, and ovarian cancer (84). Furthermore, an increase in frequency of cold and flu infections experienced also decreased risk for these cancers. We found only two studies that obtained opposite results (74, 85). Chickenpox (OR=2.09) and mumps 17 (OR=2.61) were shown to increase the risk of cancer.
As a result, Hoffman et al proclaimed that ‘no final statement could be made on the association of childhood diseases or fever and cancer should be made.’ It is difficult to compare these two studies with the majority of studies described above and come up with a reasonable explanation for the different outcome. However the majority of studies published before and after Hoffman et al. findings support the link of a history of infections to decreased cancer risk.
Not all studies examining the relationship between infections and cancer deal with solid malignancies. Increasingly reports and studies are demonstrating an increased risk for acute lymphoblastic leukemia (ALL) with decreased exposure to childhood infections.
Currently no specific pathogen has been implicated in lowering the risk of ALL development. Evidence continues to point towards a ‘delayed infection hypothesis’- that ALL risk increases with the delay in exposure to certain infections early in life. The study by Urayama et al. looked at different indicators in addition to infection in order to firmly establish that it is the early childhood infections that lower the risk of ALL.
Specifically the study examined simultaneous effects of: 1. Birth order, 2. Day-care attendance, and 3. Common childhood infections. Subjects between the ages of 1-14 were enrolled in the Northern California Childhood Leukemia Study (NCCLS) conducted from 1995 to 2008.
Approximately, 669 ALL subjects (284 non-Hispanic whites and 385 Hispanics) and 977 controls (458 non-Hispanic whites and 519 Hispanics) were selected to address the relationship to socio-demographic differences. ALL positive status was defined as a diagnosis of CD10+CD19+ALL between the ages of two to five.
Cases and controls were compared separately for each ethnicity due to inherent differences in daycare utilization and family size. The group found that non-Hispanic white children who attended daycare by six months of age (p=0.046) and who had one or more, older siblings 18 (p=0.004) had lower risk for ALL. Also both Hispanic (OR: 0.48 [0.27-0.83]) and non-Hispanic white (OR: 0.39 [0.17-0.91]) children had a decreased odds ratio for having ALL if they had an ear infection before the age of six months (86).
Additional social contact measures in Hispanic children did not demonstrate decreased risk for ALL. Other studies examining the traditional measures of childhood infections (measles, chickenpox, mumps, rubella, and pertussis) were also found to play a protective role against the development of other non-solid malignancies such AML and CLL in adulthood (87).
Anic et al. utilized similar indirect measures of infection risk with a study on adult glioma risk (88). This study examined glioma risk with birth order, family size, birth weight, season of birth, and breast-feeding history.
The study recruited 889 adult glioma patients 18 years or older that were less than 3 months from glioma resection. All glioma case participants were recruited from neurosurgery and oncology clinics from southeastern universities and cancer center. About 903 non-blood related brain tumor free individuals from the same communities were used as case controls.
Individuals were subjected to interviewer questionnaires to collect data on cancer risk. Anic et al. found that individuals who had any siblings (OR = 0.64; p = 0.020) or older siblings (OR = 0.75; p = 0.004) were at a lower risk of developing a glioma (88). All other risk factors tested were not significant.
Another study specifically dealing with Non-Hodgkin’s Lymphoma (NHL), investigated the relationship between cancer and direct (i.e. self reported infection exposure) or indirect (i.e. family size, birth order, day care attendance etc.) measures of exposure to infection.
NHL rates were rising by 3-4% in the developed world and it was hypothesized that this increase could be due to delayed infections leading to immune dysregulation (89). To test this, 1388 NHL patients, 354 Hodgkin’s Lymphoma patients, and 1718 healthy controls were recruited in Italy and 19 questioned on their family size and history of acute and chronic infectious diseases as well as autoimmune disease.
This particular study found that individuals were at an increased risk for NHL if exposure to their first bacterial or viral infection was delayed to after the age of four. Smaller family size also appeared to be greater risk factor for NHL (89).
Contracting a live pathogen is not the only way of providing protections against cancer. Vaccination with attenuated pathogens may also lower cancer risk. One group conducted a study on 542 malignant melanoma patients assessing the effect of vaccination practices on survival (90).
The particular childhood vaccinations of interest, BCG and Vaccinia, were common until the late 1970s and 1980s. Although these patients already were dealing with cancer, valuable information was gathered. According to Kaplan-Meier analysis, melanoma patients immunized as children with BCG and/or Vaccinia survived much longer than their unvaccinated counterparts.
At five years following malignant melanoma diagnosis, vaccinated group’s survival was about 75% compared to approximately 50% in the unvaccinated group. Whether BCG and Vaccinia vaccination were considered separately or jointly, the hazard ratio for death in melanoma patients was decreased compared to unvaccinated patients.
The study also analyzed the relationship between the number of reported bouts of infection and length of survival. As number of infections (that included osteomyelitis, mastitis, abscess, or furuncle) increased, the hazard ratio decreased yielding a significant difference in survival irrespective of whether the infections were accompanied by elevated temperature (p-value=0.004).
In light of the above evidence, it should be clear that we need to act with more prudence when it comes to vaccination. It’s now recommended in the United States for everyone aged six months and over to be vaccinated against influenza on a yearly basis, even though the vast majority of influenza deaths occur in people aged 65 and over. It’s questionable whether the benefits really outweigh the potential harm for many demgraphics.
I should note that although skepticism of vaccination is justified, it’s important to avoid falling into the trap that depicts vaccination as being part of some sort of conspiratorial plot against the general population. This discredits the critic and represents an unfalsifiable accusation.
A far better explanation would be to recognize vaccination as a case where people are blinded by their strong faith in their ability to control and improve upon nature. The idea that viruses might not be merely some sort of cosmological burden placed upon organisms, but actually play an important role in regulating a variety of processes in a manner that benefits their host never occurs to most people.
It’s worth noting that American culture is quite unique in its excessive faith placed upon modern medicine. There is no other country where per capita healthcare expenses are so high and where people are exposed to such a large vaccination schedule.
Having studied in the Netherlands myself, we were exposed to literature that was relatively criticial of the supposed accomplishments of modern medicine. It was explained for example, that our increase in life expectancy is largely the result of improved hygiene and nutrition.
In much of Europe, doctors tend to be more reserved and cautious in their treatment of patients. In Germany for example, St John’s wort is the most commonly used treatment for depression. In the United States on the other hand, Cymbalta is most preferred, despite the risk of suicide it carries.
Similarly, the United States accounts for 80% of global Ritalin consumption. Is something uniquely wrong with American children, or are Americans unique in their idea that every mild abberation from the norm needs to be treated with modern pharmaceutical medicine?
It’s generally accepted in polite company to raise these questions about the American preoccupation with treating every imaginable health conditions, but vaccination in contrast remains a sacred cow. If you were to suggest that there can be such a thing as “excessive vaccination”, you’re commonly interpreted as some sort of anti-intellectual.
In fact, as the evidence above would suggest, we have reason to be very skeptical of vaccination as vaccines may have caused more harm than they solved. Most of the decline in deaths from vaccine preventable illnesses occured before the vaccines were introduced. Thus, vaccines prohibited the observed effect of reduced cancer incidence after natural infection among billions of people, to prevent a very small minority of deaths that could not have been prevented through means other than the complete eradication of these viruses.
The disturbing prospect that now awaits us as vaccination gradually eliminates every childhood febrile infection is that a generation of young people is currently growing up whose immune systems are ill-prepared for the potential development of cancer later in their lives. In fact, as the study I quoted itself suggests, current cancer rates already appear to be far higher than they would be if we would simply tolerate some of the infections that regularly occur in children. Perhaps next time you see someone sneezing, you’ll say “bless you” for entirely different reasons.