Australia’s Role in Pneumococcal and Human Papillomavirus Vaccine Evaluation in Asia-Pacific

Australian researchers have made substantial contributions to the field of vaccinology over many decades. Two examples of this contribution relate to pneumococcal vaccines and the human papillomavirus (HPV) vaccine, with a focus on improving access to these vaccines in low- and lower-middle-income countries (LLMICs). Many LLMICs considering introducing one or both of these vaccines into their National Immunisation Programs face significant barriers such as cost, logistics associated with vaccine delivery. These countries also often lack the resources and expertise to undertake the necessary studies to evaluate vaccine performance. This review summarizes the role of Australia in the development and/or evaluation of pneumococcal vaccines and the HPV vaccine, including the use of alternative vaccine strategies among countries situated in the Asia-Pacific region. The outcomes of these research programs have had significant global health impacts, highlighting the importance of these vaccines in preventing pneumococcal disease as well as HPV-associated diseases.


Introduction
The pneumococcal vaccine and the human papillomavirus (HPV) vaccine are both highly successful vaccines in terms of reducing invasive and non-invasive pneumococcal disease (IPD) and HPV-associated diseases, respectively. Both vaccines are very costly, which means that many low-and lower-middle-income countries (LLMICs) are unable to afford introducing or sustaining these vaccines into their National Immunisation Programs (NIPs) without support from international organizations (i.e., PATH, the Bill and Melinda Gates Foundation, and Gavi, the Vaccine Alliance) [1,2]. Alternate vaccine schedules such as a reduced number of doses and/or extended durations between doses would alleviate cost and logistical difficulties associated with vaccine implementation in LLMICs, provided that similar vaccine efficacy and/or immunogenicity can be demonstrated between the original schedule and alternate schedules in randomised controlled trials. Evaluation of alternative dose schedules has been an important part of the global pneumococcal vaccine and HPV vaccine research agenda, in which Australian researchers have had a leading role.
Measuring vaccine impact in LLMICs once they have introduced these vaccines is also critical to inform health policies. While high-income countries typically have appropriate data collection systems and databases to monitor vaccine coverage, vaccine impact and effectiveness (including surveillance indicators), as well as vaccine safety, these systems are often lacking in many LLMICs [2]. There are many LLMICs in the Asia-Pacific region that have not yet introduced the pneumococcal vaccine and/or the HPV vaccine into their NIPs or have only just introduced them in the last 10 years. Many of these countries have limited resources and experience in monitoring the introduction and impact of new vaccines. In this review, we highlight Australia's role in measuring the impacts of the pneumococcal and HPV vaccine as well as evaluating alternative schedules among countries in the Asia-Pacific region, thus contributing to global health impact.

Burden of Disease and Pneumococcal Conjugate Vaccine
Streptococcus pneumoniae, also known as the pneumococcus, is a major cause of bacterial pneumonia in the young, the elderly, and immunocompromised individuals [3]. Between 2000 and 2015, it is estimated that >300,000 deaths in children aged 1-59 months were caused by pneumococcus [4]. Most of these deaths occurred in LLMICs due to the inaccessibility of pneumococcal vaccines, owing to their high cost, and limited access to disease treatment (i.e., oxygen and/or antimicrobial therapy) [5].
There are two types of pneumococcal vaccines available (Table 1); the pneumococcal polysaccharide vaccine (PPV) and the pneumococcal conjugate vaccine (PCV). PPV has been recommended for use in older children (>2 years old) and adults who are at increased risk of pneumococcal diseases [6], while PCV is recommended for children <2 years of age as well as in older adults. There is a long history of Australian-led research on both the use of PPV and PCV in LLMICs and high burden settings (i.e., Indigenous Australians).

Pneumococcal Polysaccharide Vaccine
In the 1970s, Australian researchers conducted important vaccine trials in Papua New Guinea (with high pneumococcal disease burden), where the findings were crucial to the subsequent licensure of the first pneumococcal vaccine (23vPPV) in 1983 [7] (Table 2). In particular, the study found that vaccination of adults with a 14-valent PPV (first generation PPV) reduced pneumococcal infection and death by 81% and 44%, respectively [8]. While PPV was effective in adults, there was little impact in children, particularly those <2 years old [9]. The low efficacy of PPV in children was attributable to the low immunogenicity of the vaccine to some serotypes in children (as a result of an immature immune system), including those serotypes that are commonly responsible for invasive diseases in children [10]. These findings, along with others, contributed to the recommendations for the use PPV in children >2 years of age.  In high pneumococcal burden settings such as Papua New Guinea and among Indigenous Australians, young infants can be colonised with pneumococcus within their first month of life, and it represents the highest risk for IPD [21][22][23]. Maternal immunisation is one way to prevent early pneumococcal carriage in young infants, and this protection is thought to be mediated through antibody transfer at the time of delivery and/or breast feeding. Lehmann et al. found that serotype-specific antibodies (5 and 23F) were significantly higher in children of immunised women than unimmunised women for up to age 2 months and for up to age 4 months for serotype 14, providing evidence that such intervention prevents early pneumococcal carriage [16]. This study, along with two other studies in Gambia [24] and Bangladesh [25], contributed to the earliest data on maternal immunisation.

Pneumococcal Conjugate Vaccine
Countries that introduced PCV, including Australia, have observed large reductions in pneumonia and IPD [4]. Of the 73 Gavi-eligible countries, 59 (81%) introduced PCV into their NIPs. In contrast, only about 50% of lower-and upper-middle income countries (non-Gavi eligible) introduced PCV into their NIPs as these countries often do not have the support from international health-care organizations, as do low-income countries [26]. Most of these countries that are yet to introduce PCV are in Asia and Africa, where disease burden is highest, and include countries with large populations, such as China and Vietnam [27]. In countries that can afford to introduce PCV into NIPs, serotype replacement is a major concern, since currently available PCVs cover only 10-13 of the >100 pneumococcal serotypes. The extent to which replacement occurs in LLMIC, where the burden of disease is the highest, is unknown. Australian researchers have contributed to evaluating the optimal pneumococcal vaccine schedules for LLMICs, as well as monitoring the impact of pneumococcal vaccines in LLMICs and in high-risk communities (e.g., the Australian Indigenous population), as discussed below.

Alternative Pneumococcal Vaccine Schedules
The cost of the PCV is a barrier for its use and sustainability [28]. A three-dose schedule is currently recommended by the WHO in children. Using fewer PCV doses, such as one primary dose with one booster dose (1 + 1 schedule), could be a more cost-effective way of using this vaccine to maintain herd protection and may improve the vaccine's financial sustainability [29]. Other strategies to improve protection in high-burden settings include maternal immunisation, as well as combining different PCVs (PCV10 and PCV13) to broaden protection against respiratory pathogens such as pneumococci and non-typeable Haemophilus influenzae (NTHi). Table 3 summarises studies in which Australian researchers have played a crucial role in evaluating alternative pneumococcal vaccination schedules for LLMICs and in other high burden settings. These studies had a major global health impact. For example, the study conducted in Fiji by Russell et al. led to the introduction of PCV into Fiji's NIP (3 + 0 schedule) in 2012 [30,31]. The study was also the first to show that a single dose of PCV7 given during infancy may offer some protection for most vaccine serotypes. Other important research relevant for high burden settings (i.e., Papua New Guinea), where pneumococcal colonisation occurs very early in life, relates to the evaluation of neonatal pneumococcal vaccination. This strategy was demonstrated to be safe and immunogenic [11,12,32]. Higher valency vaccine may also be needed in high burden settings, since infants can be colonised by multiple serotypes, including those not included in current PCVs [11,12]. An alternative strategy to increase serotype coverage until new PCVs with broader serotypes, or serotype-independent vaccines, become available, is the combination of priming with three doses of PCV and boosting with one dose of 23vPPV [33]. This strategy, however, was found to induce short-term immune hypo-responsiveness, although the clinical significance is unknown [34][35][36]. Half of each group received 23vPPV at age 12 months.
All received a small dose of 23vPPV at 17 months Three dose PCV schedule is more immunogenic than 1 or 2 doses Priming with a single dose was better than 2 or 3 doses. Less vaccine-type pneumococcal carriage with more doses of PCV Children who did not receive the 23vPPV had significantly higher antibody levels for all PCV serotypes compared with those who received 23vPPV following exposure to a small dose of 23vPPV (immune hyporesponsiveness) [30,34,[36][37][38]  To determine the impact of early schedules on pneumococcal carriage Healthy infants randomized at birth to receive PCV7 in a -0-1-2-month (n = 101) or -1-2-3-month (n = 105) or no vaccine (control, n = 106).
All children received 23vPPV at age 9 months PCV recipients had higher antibody response for all vaccine types except 6B than controls at age 3 months PPV induced significantly higher vaccine-type antibody responses in PCV7-primed infants than in controls No significant differences in PCV7 serotype carriage between PCV7 recipients and controls at any age Prevalence of non-PCV7 carriage was 17% higher in 7vPCV recipients (48%) than in controls (25%) at 9 months of age.

Pneumococcal Vaccine Impact
Evaluation of the distribution of pneumococcal serotypes causing invasive diseases and/or carriage prior to vaccine introduction is crucial for measuring vaccine impact and serotype replacement. Researchers from Australia have been involved in pneumococcal vaccine evaluation studies in the Asia-Pacific region ( Table 4). The studies in Table 4 demonstrated significant reductions in vaccine-serotype carriage and hospital admission due to pneumococcal pneumonia and/or acute lower respiratory infections (ALRIs) following the introduction of the pneumococcal vaccine, with some exceptions in Australian Indigenous cohort studies. Early effectiveness studies in Australian Indigenous infants found an increased risk of ALRIs, no change in otitis media incidence and radiologically confirmed pneumonia following three primary doses of PCV7 and a dose of 23vPPV at 18 months [48][49][50]. It was hypothesized that this limited vaccine impact could be due to early pneumococcal carriage, carriage of non-vaccine serotypes responsible for severe pneumonia, non-vaccine serotype/other respiratory pathogen replacement in the respiratory tract, as well as the immune hypo-responsiveness associated with 23vPPV. Findings from this study led to the revision of PCV immunisation schedules for Australian Indigenous infants (removal of 23vPPV at 18 months) [51]. Indirect effects on vaccine-serotype carriage in adults have been documented following PCV vaccine introduction. Not surprisingly, there has been an increase in non-vaccine serotypes in some countries (Fiji and Mongolia where data is available). The extent of serotype replacement in LLMICs, particularly countries in the Asia-Pacific region, is poorly understood, and threatens the control of pneumococcal disease. This knowledge gap highlights the need for continued surveillance and monitoring of vaccine impact in the region.  One year after introduction: PCV13 serotype carriage reduced by 52% and 51% in 12-23-month-old and 5-8-week-olds, respectively; increase in non-PCV13 serotype carriage (55%) in 12-23-month-old. Evaluation in children aged 2-59 months ongoing with the following outcomes: hospitalised radiological pneumonia, clinically-defined pneumonia, nasopharyngeal carriage of pneumococci among pneumonia patients and in the community, and severe respiratory infection associated with RSV and/or influenza. [61,62] Australian Indigenous population originally receive 3 primary doses of PCV (2, 4, 6 months) + 1 booster dose of PPV at 18 months, while non-Indigenous population receive 3 primary doses of PCV (2, 4, 6 months) with no booster. 23vPPV: 23-valent pneumococcal polysaccharide vaccine; ALRI: acute lower respiratory infection; PCV: pneumococcal conjugate vaccine.

Burden of Disease and HPV Vaccines
HPV is a broad group of viruses with more than 200 genotypes, some of which have tropism for skin, and others (~30 to 40 genotypes) for the genital mucosal and skin area [63]. Genital HPVs are transmitted by close contact, between genital skin and genital skin/mucosa, and are the most common viral sexually transmitted infection. It is estimated that approximately 80% of sexually active individuals will be infected by HPV at some stage in life, especially early after sexual debut [64]. HPV is known to cause a range of diseases from anogenital warts and benign/low-grade genital abnormalities (the viral cytopathic response) to invasive anogenital cancers, particularly cervical cancer [63]. Cervical cancer is the fourth most common cancer in women worldwide, with 604,000 cases and 342,000 deaths in 2020 [65]. There are 20 HPV genotypes that are known to cause cancer (oncogenic of high-risk types), with HPV 16 and 18 together accounting for 70% of cervical cancers worldwide [66,67].
HPV was first identified to be the causal agent of cervical cancer in the early 1980s by Harald zur Hausen and his team [68]. Since then, there has been extensive research into the prevention of cervical cancer, including the prevention of HPV infection through vaccination. The breakthrough in the development of the HPV vaccine was the discovery of the self-assembly capsid viral proteins into virus-like particles (VLPs) in Australia, and also elsewhere by others [69]. This technology subsequently became the basis for the current prophylactic HPV vaccines.
There are currently three licensed prophylactic HPV vaccines and one under review for WHO pre-qualification (Table 5). These vaccines are highly immunogenic and effective in preventing vaccine-type HPV infection, cervical pre-cancers, and cancers [70,71]. Both 4vHPV and 9vHPV are also effective against genital warts, and other vaccine-type anogenital pre-cancers such as vulvar, vaginal, and anal [72,73].

Australian HPV Vaccine Program and Impact
Australia was one of the first countries to introduce a government-funded schoolbased HPV vaccine program (4vHPV in 2007). The program was first introduced as a female only program and achieved high vaccine coverage (around 80% for three doses) in women <18 years of age. The vaccine impact was one of the first reported globally [74] (summarised in Table 6). Within the first five years of HPV vaccine introduction, significant decreases in vaccine-type HPV prevalence were observed in both men (as a result of herd protection from the female only program at the time) and women (also with herd protection of same age vaccine eligible women), as well as high-grade cervical abnormalities in women [75][76][77]. The prevalence of high-risk vaccine-type HPV declined from 22% in the pre-vaccine era to 1.5% among girls aged 18-24 years old, within nine years following introduction of the vaccine [76,78]. Cross protection against closely related HPV vaccine types (HPV 31/33/45), as represented by a decrease in HPV genotype prevalence, was also observed six years after introduction of the vaccine [77]. Seven-years post-4vHPV-introduction in Australia, a national data linkage analysis reported 40% vaccine effectiveness against high-grade cervical intraepithelial neoplasia (CIN) (all cause, non-HPV type specific, histologically confirmed) [79]. The vaccine was effective against CIN, regardless of whether women have received one, two or three doses of 4vHPV, suggesting that one dose of HPV vaccine may be sufficient for protection. Single dose HPV vaccine schedules are particularly relevant for LLMICs, where high costs and logistical difficulties in vaccine delivery are major barriers to vaccine implementation.
In 2013, Australia introduced a gender-neutral HPV vaccination program. The benefits of gender-neutral HPV vaccination include direct protection for men (including men-whohave-sex-with-men (MSM), who do not benefit from female only vaccination) and the provision of herd protection for unvaccinated women [80]. In many high-income countries where cervical cancer is controlled by vaccination and cervical cancer screening, the risk of anal cancer for MSM can be as high as the risk of cervical cancer for women [81], highlighting the importance of HPV vaccination in boys and young men. Chow et al. recently reported on a repeat cross-sectional study conducted in MSM comparing the HPV prevalence before and after the introduction of gender-neutral HPV vaccination in Australia [82]. They found a significant reduction in the prevalence of HPV genotypes 6, 11, 16, or 18 in the anus (76%), penis (52%), and oral cavity (90%) compared with a pre-vaccination cohort, demonstrating the first direct impact of HPV prevalence in MSM after the implementation of the gender-neutral HPV vaccination programme [82], which is likely to lead to reductions in anal cancer incidence.
In 2018, 9vHPV was introduced as a two-dose schedule to replace 4vHPV in Australia. It was postulated that the replacement of 4vHPV with 9vHPV in Australia will protect against an additional 15% and 11% of cervical cancer and anal cancers, respectively [74]. With the use of 9vHPV coupled with high vaccine coverage in a gender-neutral vaccination program, and robust HPV cervical screening, Australia is likely to be the first country to eliminate cervical cancer (as defined as <4 new cases per 100,000 women each year) by 2028 [83]. The incidence of cervical cancer is expected to further decrease to <1 case per 100,000 women by 2066 [83]. Indeed, completely vaccinated women in Australia were found to have less than half the incidence rate of cervical intraepithelial neoplasia grade 3 and/or cervical adenocarcinoma in Situ than in unvaccinated women (2.8 cases compared with 6.0 cases per 1000 women). A trend of lower incidence of cervical cancer in HPVvaccinated than in HPV-unvaccinated women was also reported, although longer follow up data are required to verify this observation [84].

Alternate HPV Vaccine Strategies
Licensed HPV vaccines were originally given as a three-dose schedule. In 2014, the WHO recommended a two-dose schedule (six months apart) for girls/boys <15 years old [91]. This was based on immune-bridging studies that demonstrated non-inferior antibody levels in girls <15 years old who received two doses compared to older women aged 16-26 years old who received three doses (where efficacies against HPV infection and cervical pre-cancer have been established) [92].
Australia researchers were involved in the evaluation of reduced-dose HPV vaccine schedules for Fiji and Mongolia. A cohort study in Fiji found that girls who received two doses of 4vHPV had similar immune responses, after six years, to girls who received the standard three-dose schedule [93][94][95]. This was the longest follow up of reduced-dose schedules at the time, supporting the WHO recommendation of a two-dose schedule. More interestingly, girls who received a single dose of HPV vaccine had higher antibody levels than unvaccinated girls (albeit lower levels than those who received two or three doses) after six years, and these were boosted to a similar level as girls who received two or three doses following a booster dose of 2vHPV [94]. This was the first study to demonstrate the generation of immunological memory following just one dose of HPV vaccine, as well as the first to report on the immunogenicity following a mixed vaccine schedule [94]. These findings, along with other studies on the single dose schedule [96][97][98][99], supported further research into this field [100,101], leading to several ongoing Phase III clinical trials [98,102,103]. It is, however, important to note that the clinical relevance of lower antibodies generated following one dose of HPV vaccine is unknown since there is no identified immune correlate of protection. Recent data from Mongolia which has one of the highest cervical cancer rates in Asia (age-standardised rate of 19.7/100,000) [104], reported 92% reduction in the prevalence of vaccine-type HPV 16 and 18 in girls who only received one dose of 4vHPV six years earlier compared with unvaccinated girls [105]. In addition, 90% and 58% of vaccinated women remained seropositive for HPV 16 and 18, respectively, with antibody levels significantly higher than unvaccinated women [105]. This study not only contributes to the limited HPV vaccine research in Mongolia, but also the emerging evidence of single-dose HPV vaccine schedules globally. A single dose schedule will alleviate the constraints (high vaccine costs and difficulties in vaccine delivery) faced by many LLMICs, where the burden of cervical cancer is the highest.
Another alternate strategy is to vaccinate those that are at the highest risk of HPV infection and cervical cancer, such as those with high numbers of sexual partners, in addition to cervical cancer screening. While HPV vaccines are prophylactic and do not clear existing lesions [106], it is increasingly being recognised that there are still important benefits for vaccinating HPV-infected women [107], particularly in settings where cervical cancer screening is limited (i.e., LLMICs and remote settings). These benefits include reducing transmission, protecting against vaccine-type HPV that the individual is not infected with, as well as reducing the risks of clinical disease relapse after treatment. This concept is particularly relevant for female sex workers (FSWs) who have a very high risk of HPV infection and cervical cancer, due to the high number of sexual partners [108]. It is also common for them to harbour multiple HPV genotype infections [109], potentially serving as a reservoir for transmitting HPV within the community. We are conducting a pilot study in Vietnam to investigate a targeted HPV vaccination strategy towards FSWs to reduce their risk of HPV infection and protect them against HPV-associated diseases, as well as reducing HPV transmission within the community [110].

HPV Vaccine Introduction in LLMICs
The contribution of Australian researchers to global HPV vaccine research has been fundamental to the introduction of HPV vaccines in the Asia-Pacific region. In addition, the Australian Cervical Cancer Foundation (ACCF) in partnership with either Gavi or Gardasil Access has been involved in a number of HPV vaccine demonstration projects and HPV educational programs over the past decade in LLMICs including Nepal, Bhutan, Kiribati, Vanuatu, the Solomon Islands and Papua New Guinea [111]. In particular, the ACCF has been successful in facilitating the introduction of the HPV vaccine in Bhutan, the first developing country to have a national HPV vaccination program and, more recently, in the Solomon Islands in 2019. As of October 2018, the ACCF has contributed more than 500,000 doses of HPV vaccine to these countries and has established sufficient local health authority capacity to be able to successfully undertake a complex HPV vaccination program (J. Tooma, personal communication).

Conclusions
Australia has made substantial contributions to the measurement of the impact of pneumococcal and HPV vaccines, as well as facilitating the introduction of the HPV vaccine in LLMICs. These have important implications in providing real world evidence that the vaccines are effective, and in sharing the lessons learnt both at the country level where the research was conducted, as well as for the global health communities. Ongoing research in these areas will provide the necessary justification and steps for countries that have yet to introduce pneumococcal vaccines and/or the HPV vaccine to do so, as well as to inform governments on the impact of these vaccines, so that other health funds can be directed to diseases that are not preventable. Alternative schedules that alleviate high vaccine costs and logistical constraints in vaccine delivery will improve vaccine access and reduce inequality in LLMICs. Current PCV evidence suggests that two primary doses separated by at least two months followed by a later booster dose at or after 9 months of age would provide protection for children in high burden settings, while a single-dose HPV vaccination schedule has shown encouraging results. Randomized controlled trials are ongoing to evaluate these schedules and, if successful, will significantly improve vaccine access in LLMICs that have yet to introduce these vaccines.